<<

TRADE STUDIES TOWARDS AN AUSTRALIAN INDIGENOUS SYSTEM

A thesis submitted for the degree of Master of Engineering by Gordon P. Briggs B.Sc. (Hons), M.Sc. ()

School of Engineering and Information , University College, University of New South , Academy

January 2010 Abstract

During the project landings of the mid 1970s the United States of America was the pre-eminent space faring nation followed closely by only the USSR. Since that time many other nations have realised the potential of not only for immediate financial gain in areas such as communications and earth observation but also in the strategic areas of scientific discovery, industrial development and national prestige. on the other hand has resolutely refused to participate by instituting its own space program. Successive Australian governments have preferred to obtain any required space hardware or services by purchasing off-the-shelf from foreign suppliers. This policy or attitude is a matter of frustration to those sections of the Australian technical community who believe that the nation should be participating in . In particular the provision of an indigenous that would guarantee the nation independent access to the space frontier. It would therefore appear that any launch vehicle development in Australia will be left to non- government organisations to at least define the requirements for such a vehicle and to initiate development of long-lead items for such a project. It is therefore the aim of this thesis to attempt to define some of the requirements for a nascent Australian indigenous launch vehicle system. Conceptual design studies of a capable of launching a payload of commercially viable mass into are made. The nature of a number of political and economic factors that could slow or stop such a project are pointed out and strategic choices that could minimise these effects are suggested. As a result it is concluded that the putative launch vehicle should be designed, sourced, manufactured and launched using existing Australian resources. Various parameters whose settings are given by discrete choices made such as propellant combination, propellant feed method, engine type and launch site are to be considered and an optimal choice made based on a number of selection criteria. Implications for the remainder of the design such as engine sizing and fuel tank construction where the parameters are based on continuous functions are to be pointed out and various sizing and/or costing implications made. Critical and any long lead items are to be identified. A conceptual outline of the vehicle design is presented. Finally a time line for early development items and a work breakdown structure for a following Phase-A concept feasibility study are provided.

i

ORIGINALITY STATEMENT

I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.

Signed: …………………………………………

Date: …………….…23/09/2010……………

Gordon P. Briggs

ii Acknowledgements

I wish to thank my supervisors Dr John Milthorpe and Dr Tapabrata Ray for their guidance and encouragement. I also wish to thank the University College at the Australian Defence Force Academy for the provision of a UCPRS Scholarship to allow this work to be undertaken. I also thank the staff of NASA Glenn in particular Russell W. Claus who assisted me with the use of NASA-CEA software, Lt Col John “Hondo” Gratton MBE (Retd) for commenting on much of the draft, the “WIFers” for encouragement and to Ms. Denise Russell who kindly edited the text.

iii Contents Abstract ...... i Declaration ...... ii Acknowledgements ...... iii

Chapter 1 Introduction ...... 1 1.1 Background to Australia’s Space Launch Heritage ...... 1 1.2 Motivation ...... 2 1.3 Scope of Thesis ...... 3 1.4 Major Contributions of Thesis ...... 4 1.5 Organisation of Thesis ...... 4

Chapter 2 Literature Survey ...... 7 2.1 Introduction ...... 7 2.2 Earliest Work ...... 7 2.3 The German Wartime V2 (A4) Ballistic ...... 7 2.4 NASA Launch Vehicles ...... 9 2.5 Russian Launch Vehicles ...... 11 2.6 European Launch Vehicles ...... 11 2.7 Japan, China, India ...... 18 2.7.1 Japan ...... 18 2.7.2 China ...... 19 2.7.3 India ...... 20 2.8 Scud ...... 21 2.9 Launch Vehicle Design ...... 22 2.10 Numerical Methods ...... 25

Chapter 3 System Considerations ...... 27 3.1 System Breakdown ...... 27 3.2 Basic Vehicle Configuration ...... 28 3.2.1 Launch Vehicle Mission ...... 28 3.2.2 Launch Vehicle Concept ...... 29 3.3 Launch Site Limitations ...... 31 3.3.1 Earth Rotation ...... 31 3.3.2 Trajectory Inclination ...... 32

iv 3.4 Range Safety ...... 34 3.4.1 Range Safety Constraints and Considerations ...... 34 3.4.2 Drop Zones and Azimuth ...... 36 3.4.3 Australian Space Licensing ...... 36 3.4.4 Australian Safety Code ...... 38 3.5 Guidance and Tracking ...... 40 3.6 Operations ...... 40 3.6.1 Transport ...... 41 3.6.2 Launch Campaign ...... 42 3.7 Loads and Aerodynamic Heating ...... 42 3.7.1 Maximum Dynamic Pressure ...... 42 3.7.2 Aerodynamic Heating ...... 41 3.7.3 Structural Loads ...... 42

Chapter 4 Strategic Propellant Selection ...... 44 4.1 Importance of Correct Propellant Selection ...... 44 4.2 Historical Propellants and their Problems ...... 45 4.2.1 The Good ...... 45 4.2.2 The Bad ...... 46 4.2.3 The Ugly ...... 47 4.3 Future Propellants ...... 48 4.3.1 Performance Considerations for Launch Vehicle Fuels ...... 48 4.3.2 Advanced Launch Vehicle Fuels ...... 54 4.4 Propellant Availability in Australia ...... 55 4.4.1 as Oxidiser of Choice ...... 55 4.4.2 Launch Vehicle Fuels ...... 55 4.4.3 Pressurants ...... 57 4.4.4 Spacecraft Propellants ...... 58 4.5 Conclusions ...... 58

Chapter 5 Launch Vehicle Stack Optimisation by Evolutionary Algorithm ...... 62 5.1 Stack Model ...... 62 5.1.1 Method ...... 63 5.1.2 Spreadsheet Method ...... 63 5.1.3 Velocity Budget ...... 63

v 5.1.4 Detailed Launch Vehicle Model ...... 64 5.2 Evolutionary Method ...... 64 5.3 The Calculations ...... 67 5.3.1 Computer Runs ...... 68 5.3.2 Specifying the Constraints ...... 69 5.3.3 Gradient Methods to Supplement the Evolutionary Methods ...... 70 5.3.4 Improvements to the Evolutionary Method ...... 70 5.4 Discussion ...... 72

Chapter 6 Alternative Launch Vehicle Concepts ...... 75 6.1 Launch Vehicle Velocity budget ...... 75 6.1.1 Nominal Velocity Budget Breakdown ...... 75 6.1.2 Reserve propellant ...... 76 6.2 Launch Vehicle Ideal Velocity ...... 76 6.2.1 The Equation ...... 76 6.2.2 Calculation of 44L Ideal Velocity ...... 77 6.3 Alternative Vehicle Concepts ...... 80 6.3.1 Modelling Candidate Launch Vehicles ...... 80 6.3.2 Discussion of Candidate Launch Vehicles ...... 82 6.4 Constraints on Stack Optimisation ...... 83 6.4.1 Load Factor Considerations ...... 83 6.4.2 Stage Physical Size ...... 85 6.4.3 Choice of Engine Number in First Stage ...... 92 6.4.4 The Stacks, Load Factors & Engine Commonality ...... 93 6.5 Optimisation of Two Stage Vehicle for Stage Load Factors ...... 95 6.6 Candidate Vehicle Trade Discussion ...... 98 6.7 Sensitivity of Stage Ideal Velocity to & Mass Ratio ...... 100 6.8 Sensitivity of Orbit Insertion Due to Launch Errors ...... 101

Chapter 7 Trajectory Optimisation by Evolutionary Algorithm ...... 103 7.1 Trajectory Shaping ...... 103 7.2 Computer Program ...... 103 7.2.1 Program Modes ...... 103 7.2.2 Flight Profile Model ...... 104 7.2.3 Trajectory Integration ...... 105

vi 7.2.4 Optimisation ...... 108 7.2.5 Performance of the Program ...... 108 7.3 Processing Speed Limits the Results from the Software ...... 112

Chapter 8 Launch Site and Corridors ...... 113 8.1 Introduction ...... 113 8.2 ALV-4 Launch Corridors ...... 113 8.2.1 Locating the Eastern Launch Site and Corridor ...... 115 8.2.2 Locating the Southern Launch Site and Corridor ...... 120 8.3 Summary ...... 124

Chapter 9 Synthesis of System Design Concepts ...... 125 9.1 Introduction ...... 125 9.2 The System ...... 125 9.2.1 Payload Capability ...... 126 9.2.2 Mission ...... 126 9.2.3 Launch Vehicle Concept ...... 126 9.2.4 Launch Sites ...... 129 9.2.5 Manufacture...... 130 9.2.6 Operations...... 132 9.2.7 Development ...... 135 9.3 Optimisation Software ...... 136 9.4 Recommendation Summary for an Australian Space Launch System ...... 139 9.5 Conclusions ...... 140

References ...... 141

Appendices ...... 158 A1 ROKOPT Input File ...... 159

A2 ROKOPT Output File ...... 160 B1 TRAJ2DF Input File ...... 163 B2 TRAJ2DF Output File – Flight Mode ...... 167 B3 TRAJ2DF Output File – Optimisation Mode ...... 190 C Sample STAGEX Spreadsheet ...... 201

vii List of Figures 1.1 Australia’s space launch heritage ...... 1 3.1 Conversion of Ariane AR44L to an optimised three stage launch vehicle ...... 29 3.2 ABM firing velocity geometry ...... 32 3.3 Australian space activities showing launch corridors and protected assets ...... 39 3.4 Costs of program life-cycle phases ...... 41 4.1 Theoretical vacuum specific impulse of LOX/Alcohol propellants ...... 49 4.2 Theoretical vacuum specific impulse for LOX with Alkanes and Hydrogen ...... 50 4.3 Theoretical vacuum specific impulse for LHCs and RP-1 with LOX ...... 51 4.4 Theoretical vacuum density impulse for LHCs and RP-1 with LOX ...... 53 4.5 Alternative propellant tank layouts ...... 60 5.1 Distribution of evolved GLOW population ...... 68 5.2 Distributions of ideal velocity and ideal velocity residual ...... 69 5.3 Evolved mass solutions against their evolved GLOW solutions ...... 70 5.4 Evolved GLOW solutions of a ROKOPT run using 3240 solution sets (3-D) ...... 73 5.5 Evolved stage mass solutions of a ROKOPT run using 3240 solution sets (3-D) ...... 73 6.1 Ariane 44LP Trajectory – Load Factor versus Time ...... 83 6.2 -V S-1C First Stage Structure with Tank Dimensions ...... 87 6.2 Saturn-V S-II Second Stage Structure with Tank Dimensions ...... 88 6.4 Saturn-V S-IVB Third Stage Structure with Tank Dimensions ...... 89 6.5 Payload Compartments and Fairings ...... 91 6.6 Layout of Candidate Vehicle ALV-4 ...... 99 6.7 Cases of Sensitivity to Isp and Mass Ratio ...... 101 7.1 Flight Profile with Phases, Events and Associated Parameters ...... 104 7.2 and Derived Drag Coefficient Curve ...... 106 7.3 Spreadsheet Showing Best Runs for Optimisation of the ALV-4 and Trajectory . 110 8.1 Direct Injection Trajectory resulting from a Cairns Launch showing Events ...... 116 8.2 Location of Eastern Launch Site: Overview and Locality Maps ...... 118 8.3 Parking Orbit Trajectory resulting from a Beecroft Peninsula Launch ...... 119 8.4 Location of Southern Launch Site: Overview and Locality Maps ...... 121 8.5 Approximate Trajectory for a 97 deg Sun-Synchronous Orbit Launch ...... 122 8.6 Approximate Trajectory for an ISS Inclination Orbit Launch ...... 123 9.1 Naval Air Station, HMAS Albatross at Jervis Bay (Nowra) ...... 129

viii 9.2 The Space Launch Complex 6 for the 4 Heavy LV at Vandenberg AFB .... 133 9.3 Two Boeing Delta IV First Stages Head to Horizontal Integration Facility at KSC .. 134 9.4 The Delta 4 Rocket is Lifted Upright in the MAS at Vandenberg's SLC-6 Pad ... 134 9.5 Mobile Service Tower Docked with MAS at SLC-6, Vandenberg AFB ...... 135 9.6 Timeline for First Five Years of System Development ...... 137 9.7 Work Breakdown Structure for a Phase-A study ...... 138 C1 Sample STAGEX Spreadsheet ...... 203

ix List of Tables 3.1 Launch vehicle system components and their major impacts ...... 27 3.2 Payload and equipment masses used for both AR-44L and HLV models ...... 30 3.3 HLV engine sizing data ...... 30 3.4 Earth rotational velocity at varying latitudes ...... 31 3.5 Payload required to GTO for 2681 kg delivered to GEO ...... 33 4.1 Propellant combinations and example vehicles ...... 45 4.2 Properties and performance figures of alcohol fuels ...... 48 4.3 Propellant properties ...... 51 4.4 LOX/Propellant theoretical density specific impulse as a function of mixture ratio .... 52 4.5 Cooper basin propane – quality specifications ...... 57 4.6 Trade table for propellant combinations ...... 59 5.1 Payload and equipment masses used for both AR-44L and HLV models ...... 62 5.2 Optimal stage masses as determined by STAGEX ...... 63 5.3 Stage mass ranges for the HLV as input to ROKOPT Run-A ...... 67 5.4 Values of evolutionary parameters used to evaluate the minimum GLOW ...... 68 5.5 Stage masses for the HLV as determined by ROKOPT ...... 68 6.1 AR44L Properties ...... 77 6.2 AR44L Vehicle Mass Breakdown ...... 77 6.3 Summary of Ideal Velocity Changes of Ariane 44L Launch Vehicle ...... 78 6.4 Ariane 44L Stage Mass Breakdown with Delta-V Calculation ...... 79 6.5 Parameter Selection used for Candidate Vehicles ...... 80 6.6 Historical LOX/Kerosene First Stage Properties ...... 81 6.7 Optimized Stage Masses and Lift-Off Weights for Candidate Launch Vehicles .... 82 6.8 Calculation of First Stage Burnout Load Factors ...... 84 6.9 Calculation of Off-Optimal ALV-3 First Stage Burnout Load Factor ...... 85 6.10 Stage Propellant Loading and Ratios ...... 86 6.11 Saturn-5 Stage Tank Aspect Ratios ...... 86 6.12 Candidate Vehicles First Stage Tanks Physical Size ...... 90 6.13 Candidate Vehicles Second Stage Tanks Physical Sizes ...... 92 6.14 Candidate Vehicles Third Stage Tanks Physical Sizes ...... 93 6.15 Summary of Candidate Vehicle Masses – The Stacks ...... 94 6.16 Load Factors for Engine Commonality & Structure ...... 95

x 6.17 Engine Commonality of a Two Stage Launch Vehicle ...... 97 6.18 Trade Table for Candidate Vehicles ...... 98 7.1 Optimisation Parameters for a 128 run Optimisation of ALV-4 ...... 109 7.2 Major differences between ROKOPT and TRAJ2DF ...... 111 7.3 Comparison of ROKOPT & TRAJ2DF results using calculated engine data ...... 111 8.1 Flight Data & Event Sequence for ALV-4 Vehicle using Optimised Trajectory ..... 114 9.1 Vehicle Masses for Beecroft Launch to GTO for AR44L and AR5 Equivalent.... 127 9.2 Engine Cycle Comparison ...... 127

xi List of Acronyms & Abbreviations ABM Apogee Boost Motor AIAA American Institute of Aeronautics and ALVn Alternative (Candidate) Launch Vehicle (concept) number “n” AMF Apogee Motor Firing AO5 Australis-Oscar 5 (Melbourne University Astronautical Society Hamsat) AOCS Attitude and Orbit Control Subsystem AR Ariane (launch vehicle) European Space Launch Services Company, operator of Ariane ASRI Australian Institute CEA Chemical Equilibrium with Applications (a NASA computer program) CEP Circular Error Probability

CH4 Methane C of G Centre of Gravity also CoG CRCSS Cooperative Research Centre for Systems CSG Centre Spatial Guyanais, the French/European launch site at . CYIS Cape York International ∆v Delta-V, a velocity change DO Drift Orbit (relates to geostationary ) DSTO (Australian) Defence Science and Technology Organisation ε Nozzle expansion ratio, or in context, a small amount, a residual EA Evolutionary Algorithm(s), also GA ELDO European Launcher Development Organisation ELV Expendable Launch Vehicle ESA ESRO European Space Research Organisation GA Genetic Algorithm(s), also EA GEO Geostationary Earth Orbit, (also GSO) GLOW Gross Lift-Off Weight GSO alternatively, GTO Geostationary Transfer Orbit

H2O2 HC Hydrocarbon

xii HLV Hypothetical Launch Vehicle (the baseline launch vehicle concept) HTP High Test Peroxide, concentrated (85%+) hydrogen peroxide

Id Specific Density Impulse

Isp Specific Impulse (at a given ambient exit pressure/altitude)

Ivac Vacuum Specific Impulse ICBM Intercontinental IRBM Intermediate Range Ballistic Missile IRFNA Inhibited ITAR US International Trade in Arms Regulations JBIS Journal of the British Interplanetary Society KSC Kennedy Space Centre (USA)

L* Lstar, The specific length of a combustion chamber, a measure of ignition and combustion delay of the propellant. L/D Lift/Drag ratio or in context Length/Diameter ratio (aspect ratio) LEO

LH2 Liquid Hydrogen LHC Light Hydrocarbons LNG Liquefied Natural Gas (mainly Methane) LOX Liquid Oxygen LPG Liquid Petroleum Gas (mainly Propane) LPRE Liquid Propulsion Rocket Engine LV Launch Vehicle m1, m2, m3 First, second and third stage masses respectively

N2H4

N2O Nitrous Oxide

N2O4 Tetroxide, also NTO NASA National Aeronautics and Space Administration (USA) NM Nelder and Mead multi-dimensional point improvement simplex minimisation method (Amoeba) NTO Nitrogen Tetroxide MET Mission Elapsed Time MMH Mono-Methyl Hydrazine MRBM Medium Range Ballistic Missile

xiii MTCR Missile Technology Control Regime MUAS Melbourne University Astronautical Society POGO A low frequency (e.g. 16 Hz) longitudinal oscillation of a rocket airframe REAP Engine Advancement Program RFNA Red Fuming Nitric Acid RP1 Number 1 RP2 Rocket Propellant Number 2 S1, S2, S3 First, second and third stages respectively SG Specific Gravity SL Sea Level SLV Space Launch Vehicle SSTO Single Stage to Orbit UDMH Unsymmetrical Di-Methyl Hydrazine UH25 A mix of UDMH and 25% Hydrazine Vac Vacuum VEB Vehicle Equipment Bay WFNA White Fuming Nitric Acid

xiv Chapter 1 Introduction

1.1 BACKGROUND TO AUSTRALIA’S SPACE LAUNCH HERITAGE In his 1962 book, Southall quotes the US National Aeronautics and Space Administration (NASA) as predicting that “…by 1963 the Woomera rocket range will be the finest space- research station of its type in the world” [Refs 1.1, 1.2]. Indeed, at that time the activity at the range was indeed considerable. The range, located in South Australia, north-west of , had been constructed beginning in 1946, as a long range weapons testing facility. As well as the strictly military weapons development and testing projects carried out at the range there were a number of upper atmosphere science rocket projects such as the British and the Australian Long-Tom.

Figure 1.1 Australia’s Space Launch Heritage Left: being erected for launch, Right: Main Body being erected into the launch gantry. Figures taken from Ref 1.1

The British long range weapons themselves constituted the basis of a space launch capability. Firstly the Black Knight (Figure 1.1), designed as a re-entry investigator and the Blue Streak (Figure 1.2) as an IRBM which would have become the first stage of a European three stage launch vehicle, . Equipped with a high energy upper stage Blue Streak would have been the equivalent of a US - launch vehicle. Finally in 1975 the British Black vehicle built from the technology of Black Knight orbited the British Prospero satellite.

1

Chapter 1 Introduction

A number of spaceflight activities within Australia have recommenced since that time including hypersonic propulsion activities at the University of Queensland and at the University of New South Wales at ADFA1. The Defence Science and Technology Organisation (DSTO), the successor to WRE, has again taken an interest in hypersonic propulsion and since 2006 has been funding a consortium of universities in the HyShot, HyCAUSE and HIFIRE programs [Refs 1.5, 1.6]. HyCAUSE flew in 2007, and HIFIRE 1 and 2 have also flown successfully. Under Round 1 of its “Australian Space Research Program” the Australian government, under the Auspices of the newly formed unit, has recently funded the Scramjet-Based Access-To-Space Systems (Scramspace) project. Woomera is also used as a launch site for various foreign launch campaigns keeping the range open at a steadily increasing activity level such that it is now difficult to obtain a launch window. Four generations of the AUSSAT/OPTUS communications spacecraft system have been procured and operated since 1979 [Ref 1.7]. Numerous spaceport studies have also been carried out, among them the 1966 proposal for an ELDO launch site at Darwin [Ref 1.8] and the 1986 proposal for an international launch site on Cape York peninsula in Queensland [Refs 1.9-1.14]. However no space launch vehicle project has been commenced and it has become apparent that no Australian government of any political persuasion, past or present, intends to provide the impetus to provide Australia with a capability that would guarantee an independent national access to space or to fund any projects involving spaceflight. Even the recent spacecraft project, FEDSAT, [Ref 1.15] was not funded sufficiently to provide for a launch. The project accepted the generous Federation celebration gift of a 2003 launch from the Japanese government. Small solid propellant launch vehicle projects such as the September 1987 Project Capricorn [Ref 1.16] have been proposed but have not proceeded past the concept stage. The Capricorn ALLV2 was intended to be the first customer for the Cape York International Spaceport proposed to be built in northern Queensland. In concept Capricorn was to provide a launch service of up to 1 tonne to LEO3 for the scientific community both Australian and international. It was to be manufactured entirely in Australia. While the Queensland government showed interest in having a proposed launch from the putative Cape York spaceport, no interest in the launch vehicle itself as a project was shown by any Australian industrial organisation or government. The Capricorn proposal was followed by the Southern Launch Vehicle project [Refs 1.17-1.20] which proposed to use solid stages from surplus US military to build the stack of a launch vehicle to provide similar (~750kg to LEO) launch missions as had Capricorn. In the event, the Orbital Sciences corporation of the United States successfully designed and constructed the Taurus and series of solid propellant launch vehicles [Refs 1.21, 1.22] to fill the niche that could have been occupied by the Capricorn and Southern Launch Vehicle, so that even without the lack of interest shown by Australian industry and government the opportunity was lost for Australia to develop a small launch vehicle.

1 ADFA – Australian Defence Force Academy 2 ALLV – Australian Light Launch Vehicle 3 LEO – Low Earth Orbit

2

Chapter 1 Introduction

1.2 MOTIVATION Although there is no official interest in spaceflight or independent access to space within Australia there still exists a sense of frustration and an undercurrent of interest amongst the engineering and science communities. This is expressed as a desired to participate in the design, construction and operation of facilities and flight vehicles amongst engineers and in flying scientific, experimental and exploration payloads amongst the scientists. An example of this is given by the currently ongoing project run by ASRI4, an independent public research institute, to construct a small liquid propellant launch vehicle [Refs 1.23, 1.24], intended initially as a and later to grow to a small orbital launch vehicle. ASRI also operates a program of very small solid propellant sounding to carry payloads for school projects [Ref 1.25]. It is taken as a fundamental that without a launch vehicle system a nation does not have its own space program but is always dependent on the willingness of other nations to cooperate as given by the example of the launch of FEDSAT. Given the lack of support from the Australian Government it would seem therefore that it is up to non-government organisations such as Universities, ASRI and other private companies and organisations to point the way towards an independent Australian access to space by carrying out the studies and initial research preparatory to starting a launch vehicle project.

1.3 SCOPE OF THESIS It is the aim of this thesis to carry out some of the trade studies that would point the way for organisations wishing to start the development of long lead items required for a full scale Australian space launch program. It is important to consider the entire system rather than just the launch vehicle. The whole of the lifecycle costs of the launch system must be considered if the system is to be commercially competitive. This thesis attempts to examine some of the factors that will aid in the optimisation of the total project rather than just the vehicle in order to select technologies and design variables appropriate to local conditions in the Australian environment. If a space launch vehicle is to be built by Australia it will be constrained by the capabilities of Australian industry and the potential launch sites afforded by Australian territory. The geographical location of Australia means that latitudes of possible Australian launch sites are not necessarily higher than (KSC) and are lower than Baikonur5 but industrial, infrastructure and other constraints may require launch to be from a higher latitude than KSC. The industrial capacity of Australia, while advanced, is not as varied as that of US, Europe, Japan or . The limited capabilities may require use of what might be considered non-optimal technologies in the construction of the vehicle. These may be different from those appropriate to overseas projects. The whole of the system, starting with the design mission, is examined to determine the major constraints. As the launch vehicle is the most technically complicated of the elements of a launch system, most attention is paid to that element of the system. In order to determine the most appropriate settings for a further design study. It is shown that most of the lifecycle costs occur in the operational phase and that it is here that the major savings can be made. Not all variables are open to optimisation however.

4 ASRI – Australian Space Research Institute 5 – Launch site in Kyzylorda province of Kazakhstan, rented and operated by Russia.

3

Chapter 1 Introduction

For example the location of the launch site is severely constrained by the availability of suitable land. The thesis does not attempt to come up with a system design but rather to recommend the general direction that should be followed, and the constraints and goals that must be borne in mind during the design of the system. This thesis should be regarded as a pre phase-A trade study rather than concept design. The launch vehicle is sized by a number of methods, the goal being to develop software to utilise evolutionary algorithms to carry out vehicle design. The optimisation of the launch vehicle stack is optimised initially followed by optimisation of the stack and trajectory together with a number of constraints included. For example the ability to push or pull fallout zones for the stages is one of the desirable constraints.

1.4 MAJOR CONTRIBUTIONS OF THESIS The thesis has four major contributions: 1. The thesis carries out trade studies to demonstrate the path to be followed in the development of an Australian space launch vehicle system. 2. The applicability of evolutionary algorithms to the optimisation of a launch vehicle stack model as a preliminary to a full vehicle optimisation is demonstrated. 3. Evolutionary algorithms are utilised in a computer program to minimise the lift-off weight of a launch vehicle by optimising the stack and trajectory together. The effectiveness of the method to produce flyable trajectories is demonstrated. 4. A synthesis of concepts to form a commercially viable Australian launch vehicle system is presented along with a timeline for early developmental work. A work breakdown study for a Phase-A concept design and feasibility study is also provided.

1.5 ORGANISATION OF THESIS The thesis is organised into nine chapters. This first chapter looks at the motivation and aims of the work while the second chapter carries out a review of relevant literature to basic launch vehicle design. The review is organised as a history of launch vehicle development from the early theoretical work to the German World War II V-2 missile through the Anglo-Australian joint project, the US NASA and Russian launch vehicles to the French post war work and the establishment of ESA and the development of the Ariane series of launch vehicles. This is followed by a short description of the space programs and launch vehicles of Japan, China and India. The launch vehicle description is ended by a section on the which evolved from the V-2. The above sections are then followed by a review of works on Launch vehicle design: Propulsion, Engineering, Propellants, and Operations. The chapter ends with a review of useful literature regarding general numerical methods and in particular evolutionary algorithms and their use in engineering. The third chapter examines the system considerations that must be taken into account. These include baseline mission, basic vehicle concept, earth rotation and latitude of the

4

Chapter 1 Introduction launch site, tracking and telemetry requirements, range safety, licensing and operations. The chapter concludes with a brief mention of loads and aerodynamic heating. Chapter four examines propellant choice. As an ongoing consumable, correct propellant choice continues throughout the entire lifecycle of the launch system. A wrong choice cannot be rectified without major changes to the entire system. Major considerations are performance, toxicity of propellants and exhaust, availability and cost. The chapter starts by discussing the above factors then continues by looking at historical propellants and their problems and continues by discussing future propellants. The NASA CEA program is used to calculate and compare performance values for various propellant combinations firstly of classical and near future propellants then of advanced launch vehicle fuels. The chapter concludes by examining propellant availability in Australia and then a trade discussion between the various choices. Chapter five starts the development of a vehicle (and system) optimisation utilising EA methods by examining the characteristic velocity and structural mass breakdown of the Ariane 44L vehicle. The Ariane-44L vehicle is the largest of the to 4 series of launchers and can be equated to the architecture recommended as a baseline vehicle concept in chapter three. Ariane is chosen as the archetype to reverse engineer because of its outstanding success as a commercial launcher over several decades. The Ariane-5 vehicle has a greater payload capability but its architecture of a single liquid propellant core utilising solid propellant strap-on boosters is against the recommended architecture explained in chapter three. A stack model is built up and modelled on a spreadsheet and duplicated in a mixed language computer program (ROKOPT) using FORTRAN and C++. ROKOPT optimises the mass distribution between stages utilising EA to demonstrate the feasibility of the approach. A discussion of the performance of EA in this problem and its drawbacks and advantages rounds out the chapter. Chapter six commences with a calculation of the Ariane 44L characteristic velocity, a table demonstrates the method of calculation. A number of alternative launch vehicle models are then presented. None of these alternatives utilise strap-on boosters and all are liquid propellant. The variations between the alternatives are in the different propellants used. Structural data is taken from historical vehicles mainly from Mark Wade’s excellent web resource “Encyclopedia Astronautica.” The five alternatives are compared to determine physical size constrained by load factors and the desire to have engine commonality between stages one and two to minimise engine development. A trade discussion is carried out between the competing vehicles and a diagram of the basic architecture of the preferred all LOX/LH2 vehicle is presented. Chapter seven extends the use of the EA method utilised in chapter five to optimise the stack and the trajectory together. A computer program (TRAJ2DF) is presented to carry out the optimisation. TRAJ2DF has two modes of operation: Optimise and Flight. A flight profile model is presented and linear tangent steering is explained. The method of integration along the trajectory and the aerodynamic drag model used is discussed. The objective function and constraints are discussed followed by a discussion of the performance of the program. It is concluded that while TRAJ2DF operates satisfactorily and gives flyable trajectories, its processing speed was too slow to allow a sufficiently large population to adequately fill the variable space; consequently the program should in the future be implemented on a multiple CPU facility. Chapter eight discusses possible Australian launch corridors and launch sites. It commences with a tabulation of the flight data and event sequence for the preferred ALV-4 launch vehicle

5

Chapter 1 Introduction candidate as determined by TRAJ2DF. A discussion regarding the location of a launch site for easterly launches (LEO and GTO) and then for southerly (Polar) launches follows. The thesis concludes with chapter nine forming a synthesis of concepts discussed in the preceding chapters. The various elements of the system from design, development, manufacture and operation are discussed. A time line for early development work is presented along with a work breakdown structure for a Phase-A (concept feasibility design) study. A recommendation summary for an Australian space launch system is presented.

6 Chapter 2 Literature Review

2.1 INTRODUCTION This chapter attempts to carry out a literature review of relevant documents that will give the reader an overview of launch vehicle technology. It is not possible to cover the entire literature of spaceflight due to its enormous size. However selected material is quoted which will give an overview of the development of launch vehicles from their beginning with particular emphasis on aspects that could influence the conceptual design of an Australian space launch system. Much of the material quoted is from popularly written works as these give a much greater breadth of information than technical papers from engineering journals. A number of resources are also listed. These are however constantly being updated and some are only valid at the time of writing as they can be removed from the web at any time.

2.2 EARLIEST WORK Several pioneer workers laid the theoretical and experimental bases for rocket propulsion. First was the Russian Konstantin Eduadovich Tsiolkovsky with his publication “The Exploration of Cosmic Space by Means of Reaction Devices” [Ref 2.1] in which he developed the rocket equation which determined the velocity achievable by a rocket motor or engine relating the weights of the rocket and the performance of the propellant. The second worker was the American Professor at Clark University, Worcester, Massachusetts, Robert Goddard who published a Smithsonian institution report in 1919 entitled “A Method of Reaching Extreme Altitudes” [Ref 2.2]. Goddard also carried out practical development of rocket engines and launched the first liquid fuelled rocket in March 1922 [Ref 2.3] at Roswell, New Mexico. The third pioneer was the Romanian, who published “Die Rakete zu den Planetenraumen” (By Rocket into Planetary Space) in 1923 [Ref 2.4]. In 87 pages he covered almost all major aspects of spaceflight including a design for a two stage rocket utilising LOX/Alcohol-Water propellant for the first stage and LOX/LH2 for the second stage. In 1929 he published an expanded 429 page version of the book with greater detail titled “Wege zur Raumschiffahrt” (Ways to Spaceflight) which included descriptions of the ion engine and electric propulsion [Ref 2.5].

2.3 THE GERMAN WARTIME V2 (A4) BALLISTIC MISSILE The German “Vergeltungswaffe Zwei” (“Vengeance Weapon 2”) or V-2 known to its designers as Aggregate-4 (Assembly-4) or A-4 was the ancestor of the ballistic missile as well as today’s rockets used for . Its designers, notably Dr , were originally drawn from the German rocket society, the VfR (Verein für Raumschiffahrt) and were originally inspired by the dream of rocket propulsion as a means of achieving spaceflight. The V-2 design and development team was based at Peenemünde in under the auspices of a German army team commanded by Major General Engineer Doctor . The V-2 development story was described by Dornberger in his book Der Schuss ins Weltall (The Shot into Space) [Ref 2.6].

7

Chapter 2 Literature Review

The discovery of the V-2 development site at Peenemünde by British intelligence using aerial reconnaissance and the subsequent bombing is covered in “Rocket” by Air chief Marshall Sir Phillip Joubert [Ref 2.7]. King and Kutta [Ref 2.8] cover similar ground including the V-1 (Buzz bomb) development and allied countermeasures and the V-2 attacks on Antwerp (Impact – The History of Germany’s V-weapons in World War II). King and Kutta include an interesting chapter assessing the value of the V-1 and V-2 in fighting the Second World War and of the V-2 descendent, the Scud, in the first Iraqi war. They also provide an extensive reference list of 348 references a bibliography of reports, official histories, memoirs and first person accounts, books and periodicals. The early achievements of the V-2 engineers are covered from a personal point of view by Ordway and Sharpe in “The Rocket Team” [Ref 2.9]. Their histories are followed from their early days prior to Peenemünde to their participation in the Saturn Moon Rocket program. Kraemer, in his book “Rocketdyne: Powering Humans into Space” [Ref 2.10], provides a chapter, “Rockets, from Theory to the V-2”, summarising the early history of the V-2. He gives some interesting facts re the design choices of the V-2, e,g, that the turbine driven propellant pumps came from the light weight centrifugal pumps used by firefighters. The race between the British and American forces to capture German V-2 equipment and personnel at the end of the war before the is covered by McGovern in his book “Crossbow and Overcast” [Ref 2.11]. Many unfired V-2 vehicles went to the United States for test firing and some to the British who carried out an extensive reverse engineering examination and flight testing program near Cuxhaven utilising surrendered German technicians and operational crews. The British test program was called BACKFIRE and an extensive technical and operational report was prepared [Ref 2.12] along with documentary films of the operational assembly, deployment and firing of the missile. This report remains one of the most extensive technical documents on the V-2 missile. A copy is held in the Australian Defence Force Academy library. According to the web resource, Wikipedia, a V-2 survives at the Australian War Memorial, Canberra, including a complete transporter. Wikipedia claims that this rocket has the most complete set of guidance components of all surviving A-4s with the Meillerwagen being the most complete of the three examples known to exist. Another A-4 was on display at the RAAF1 Museum at Point Cook outside Melbourne but now also resides in Canberra [Ref 2.13]. A web resource on spaceflight by Mark Wade also includes extensive information and references on the V-2 [Ref 2.14]. A more recent publication by Dungan, “V-2: A Combat History of the First Ballistic Missile” [Ref 2.15], assembles an extensive collection of original documents, unpublished photographs and accounts from those persons involved. Dungan provides a complete description of the V-2 program, the missile’s use in combat, and the race to capture its secrets. Dungan is a leading authority on the V-2 rocket and founding member of the International V-2 Rocket Research Group and assisted with the restoration of the V-2 belonging to the USAAF Museum. The above references are but a small sample of the information available on the V-2 missile. A web search on the missile will find the reader overwhelmed by the amount of information available.

1 RAAF – Royal Australian Air Force

8

Chapter 2 Literature Review

2.4 NASA APOLLO PROGRAM LAUNCH VEHICLES The NASA launch vehicles used in the development of the Apollo program were based on ballistic missiles. The Mercury program used the Redstone MRBM for the sub orbital of Alan Sheppard and Gus Grissom. The orbital flights of John Glenn, Scott Carpenter, Wally Schirra and Gordon Cooper utilised the Atlas ballistic missile. The Gemini program utilised the II ballistic missile. All launch vehicles were heavily modified from the military version to be man-rated. Only in the Apollo flights themselves was the Saturn launch vehicle specially designed for the program.

Mercury-Redstone: After the German engineers had been collected by OPERATION OVERCAST (later renamed ) US Army Ordnance moved its 130 captured staff from Fort Bliss at El Paso to its Redstone Arsenal at Huntsville Alabama. Wernher von Braun became head of guided missile development division at Redstone. The Redstone medium range ballistic missile (MRBM) was one result of work at the arsenal. The Redstone was a direct descendant of the German V-2 with a number of improvements notably a new less complicated engine with a single injector head instead of a collection of injector cups and initially still used LOX/Ethanol-Water as propellant. Additionally, the airframe was cylindrical instead of curved. The vehicle was still steered by graphite vanes in the exhaust as was the V-2. Mark Wade’s Encyclopedia Astronautica [Ref 2.16] gives extensive summary of details of the Redstone including development history, chronology, specifications and costs. Redstone modifications to the Juno, Jupiter, and Mercury-Redstone vehicles are also given. Svenson, Grimwood and Alexander [Ref 2.17] give a detailed history of the development program to develop the Mercury-Redstone vehicle including flight trials with the capsule attached. Grimwood gives a detailed chronology of the Mercury-Redstone program in Ref 2.18 There does not appear to be a text written specifically about the Redstone but there are numerous web pages devoted to the vehicle. One of the most detailed documents available is a Marshall Space Flight Center (MSFC) document [Ref 2.19] covering the Mercury-Redstone (M-R) project. It includes: vehicle description, man-rating, development test program, checkout and launch operations and the flight test program. Also included is a large bibliography of technical documents related to the M-R project.

Mercury-Atlas (MA): John Chapman of Convair Astronautics has written a detailed history of the development and flight testing of the Atlas from its MX-774 test vehicle through to its deployment in underground silos [Ref 2.20]. Chuck Walker, who was the manager of program control for the Atlas project describes the story of the Atlas rocket, in its roles as both a civilian and a military vehicle [Ref 2.21]. The book includes technical details of the rocket, including its origins as the MX-774 prototype missile, the development and deployment of its nuclear payload, its part in the Strategic Air Command squadrons, and previously unpublished pictures. The missile's conversion to a civilian rocket is also documented, including details on its role in the manned and its later use with the Agena and high- performance Centaur rocket stages used for space launches. Web resource Wikipedia gives an overview of the Mercury program including the MA flights using the Atlas as launcher [Ref 2.22]. Reference 2.23 gives flight results for the fourth manned orbital flight in Project Mercury (carrying Gordon Cooper) and summarises the entire Mercury program. It includes data for the performance of the Atlas boosters

9

Chapter 2 Literature Review used. More Atlas flight performance details can be found through the NASA Project Mercury home page [Ref 2.24] and in Refs 2.17 & 2.18. Extensive details of the development of the liquid propellant engines of both the Redstone and Atlas (and Navaho) can be found in Sutton’s History of Liquid Propellant Rocket Engines [Ref 2.25]. Again, a web search will turn up many references to the ATLAS missile, some with more or less technical detail.

Gemini-Titan II: The Titan II was an entirely different vehicle in that it used Hypergolic propellants (Nitrogen Tetroxide/: a mix of 50% UDMH and 50% Hydazine). Sutton [Ref 2.25] describes problems in the development of the Titan engines including the POGO problems. Stumpf [Ref 2.26] has written an official history of the Titan program including a development history. He covers some of the failures including those due to failure of second stage ignition, those due to combustion instability as well as some of the fatal accidents with the vehicle due to explosions in the silos. Stumpf includes several hundred references regarding all aspects of the operational lifecycles of the siloed missiles, Atlas, Titan I and Titan II. The web resource Wikipedia has several pages and references devoted to the Titan family of launch vehicles [Ref 2.27, 2.28] as does the Mark Wade Resource Encyclopedia Astronautica [Ref 2.29], The NASA history publication “On the Shoulders of Titans” [Ref 2.30] gives a history of the Gemini program mainly regarding the spacecraft and the flight program however a section is included on the problems with the Titan-II as the for Gemini. Problems encountered in proving and man-rating the booster included guidance problems, POGO and combustion instability in the second stage engine which resulted in Gemsip (Gemini Stability Improvement Program). Ref 2.31, “: A Chronology”, also gives a diarised account of the Gemini program including the booster development.

The Saturn Family: Stages to Saturn by Bilstein [Ref 2.32] gives a detailed account of the development of the Saturn Booster for the Apollo program. Of considerable interest is the section on the Saturn-1C vehicle developed by clustering Juno vehicle tankage together in what became known as “Cluster’s Last Stand”. The concept ran into a number of “gotchas” such as equalising the feed rate from the separate clustered tanks. Ref 2.33 from Marshall Space Flight Centre gives a detailed technical description of the Saturn vehicles. Apart from the references above there does not appear to be any text dedicated to the Saturn launch vehicle family. However detailed descriptions of each of the flight vehicles can be found in their NASA Apollo mission flight manuals. For example see Reference 2.34. A much more comprehensive description though can be found in the “Apollo News Reference”. This is a 128 page document which describes the history and rationale of development of the Saturn family, a first stage fact sheet, an F-1 engine fact sheet, a second stage fact sheet, a third stage fact sheet, a J-2 engine fact sheet, an instrument unit fact sheet, description of manufacturing methods and test facilities, description of testing, documentation, assembly and launch, program management and personnel. Also included is a glossary of terms and a listing of all Saturn contractors. Parts 1 to 10 of this document can be found at the location of reference 2.35.

10

Chapter 2 Literature Review

2.5 RUSSIAN LAUNCH VEHICLES The Russian launch vehicle, originally the R-7, dates back to the start of the Russian space program and is still in use today. A version has been adopted by Europe (ESA) as one of its commercial launch vehicle for payloads too small for the Ariane launcher. Reference 2.36 describes the European launch site facilities for this Russian launcher at CSG, Kourou, . A complete history of the Soyuz launcher is given by Hall and Shayler in Reference 2.37 “Soyuz: A Universal Spacecraft”. They cover the origins of the R-7 launch vehicle and its spacecraft and the many variants including the docking missions (1966-1970), Soyuz Ferry (1971-81), Progress (1978-), Soyuz-T (1979-1986), Soyuz-TM(1986-2002), Soyuz-TMA (2002). Hall and Shayler give many references regarding the Soyuz type vehicles but most are in the . Until recently a web resource “Uncle Rocket” was available from which a CD could be ordered containing drawings of the entire R-7 rocket system including details such as schematics of the turbopumps and combustion chambers. The resource has recently been removed but the CD is still extant. In Encyclopedia Astronautica, Wade also gives a detailed history of the R-7 in its many forms (such as Soyuz and ). Manufacturing history, missions and performance are included. Wade also gives data for the second most well known Russian launcher, the . Proton, intended for the Soviet manned lunar mission, is fuelled by N2O4/UDMH on the first, second and third stages and LOX/Kerosene in the fourth stage. Soyuz (R-7) is fuelled by LOX/Kerosene. Wade’s resource also gives many references but again most are in the Russian language. Baker [Ref 2.39] gives a history of “The Rocket” including the Soyuz and other Russian and other international launch vehicles. He also includes a detailed history of the Russian ICBMs. Baker’s second work “The History of Manned Spaceflight” [Ref 2.40] enlarges upon the subject in the area of manned space flight. Baker also covers the US manned space program. Clark [Ref 2.41] discusses the use of the R-7 in the Soviet manned space program. Isakowitz et al [Ref 2.42] give technical descriptions, availability, performance and cost figures for the world’s launch vehicles. The Russian LVs included are , , Proton, Rockot, Shtil, Start, , Soyuz and . Related Ukrainian LVs listed are Denpr, Tsiklon and .

2.6 EUROPEAN LAUNCH VEHICLES The Anglo-Australian Joint Project and Later: The story of British and Australian launch vehicles can be told within the context of the Anglo-Australian joint project story. In 1946, in the shadow of the German V-weapons, the Australian government under Chifley joined in a partnership with the British government of Clement Atlee to create an experimental guided weapons range at Woomera and a research and development facility near Adelaide at Salisbury. It was here that the development of many guided weapons was carried out. Australia’s contribution to the joint project was the provision and operation of the range and instrumentation. Britain shared part of the cost of operating the range and developed the long range weapons.

11

Chapter 2 Literature Review

It was here that liquid fuelled rocket such as Black Knight, and Blue Streak were launched. Australia’s first spacecraft was launched from Woomera on a US Redstone launch vehicle during project Sparta. Morton [Ref 2.43] covers the history of the Long Range Weapons Establishment from its beginnings to the end of the joint project. He includes discussions about many of the weapons and launch vehicles that operated from the range. He includes chapters on early 2 missile trials, instrumentation, life in the village of Woomera, JINDIVIK [Ref 2.44] and 3 SKYLARK as well as Australian designed and built equivalents, e.g. LONG TOM. Separate chapters are provided for the space launch vehicles Black Knight, Blue Streak and its reincarnation as the ELDO4 Europa first stage. Chapter 24 covers the development of Australia’s first spacecraft WRESAT and Chapter 25 the British launcher: Black Arrow. Southall [Ref 2.45] reports on his visit to Woomera in 1962 in a popular style. He follows the checkout, countdown and launch of a Black Knight vehicle as well as discussing facilities on the range. He also gives an overview of the living conditions for the staff on the range. Blue Streak as a weapon is also covered. Southall’s second book [Ref 2.46] on the subject is written for junior readers. Sutton [Ref 2.47] includes a chapter on Liquid Propellant Rocket Engines in the or Britain. He discusses early British work following World War 2: PROJECT BACKFIRE, development of JATO units and development of Hydrogen Peroxide engines based on the German Walter technology. These engines evolved to the engines for Blue Steel5 and to the Gamma engines for Black Knight and Black Arrow. Sutton discusses some of the development of the Blue Streak RZ-2 engine from the licensed US Rocketdyne engine. He concludes the chapter by discussing the development of small thrusters by Royal Ordnance plc. The History of the UK Rocket and Space programme from 1950-1971 is discussed in detail by Hill in Reference 2.48. He generally covers the same material as does Sutton but in much greater detail. After Blue Streak was cancelled as a weapon it continued life for a while as the first stage of the ELDO Europa launch vehicle. Hill also discusses some of the proposed developments of Black Knight and Blue Streak into a British space launcher as well as an upgraded ELDO Europa launcher. These proposals never bore fruit due to the difficulty of economically designing a capable launcher based around Blue Streak without a major redesign and redevelopment. Consequently the British Government lost interest and funding was not forthcoming. Encyclopedia Astronautica [Refs 2.49 – 2.53] has entries for the missile but none for its Stentor engine. Entries are included for Black Knight, Blue Streak and small entries for the Gamma and RZ-2 engines. The entries for Blue Streak and the RZ-2 engine appear to be at least partly incorrect or confused as Blue Streak was not manufactured by ELDO but by Hawker de Havilland. RZ-2 was a two chambered engine with two turbopumps as shown in Sutton [Ref 2.47 p860] not one chamber as shown by Wade. Wynn [Ref 2.54] gives a comprehensive discussion of the development, deployment and operation of Blue Steel and Blue Streak. Europe’s space launcher program started with the ELDO Europa vehicle based on the Blue Streak first stage. Australia was an part of ELDO supplying the launch range and

2 Jindivik: “The Hunted One”, was an unmanned target aircraft 3 Skylark: A British Sounding Rocket 4 ELDO: European Launcher Development Organisation 5 Blue Steel: An air launched stand-off bomb

12

Chapter 2 Literature Review instrumentation [Ref 2.43]. Successive failures of the Europa vehicle, the poor location of Woomera for geostationary launches and the difficulty of converting Europa based on a Blue Streak first stage to launch sufficiently large spacecraft [Ref 2.48] led to the cancellation of Europa and ELDO. Australia then lost its membership of the space launcher club [Ref 2.43]. This was never regained. Harvey traces the development of the European space program [Ref 2.55] to “…Ariane and Beyond”. He starts with the German V-2 leading to the British Backfire project then devotes chapter two to summarise Blue Streak, ELDO, ESRO6, Europa I and Europa II (Europa utilising Blue Streak after Britain withdrew from ELDO) and Europa III proposals. He ends the chapter with the ignominious end for the Europa II F-12 vehicle being used as a chicken coop after the program was cancelled. The removal of ELDO from Australia and the withdrawal of Britain from ELDO effectively ended Anglo-Australian participation in launch vehicle programs for many years. Britain took only a small financial interest in the upcoming Ariane program. Although still carrying out weapons trials at a reduced rate, Australia had no launcher activity apart from using the Woomera range to launch the British Black Arrow launcher carrying the Prospero spacecraft in October 1971 [Ref 2.55 pp83-90]. Chapter 26 of Morton [Ref 2.43] outlines the eventual run- down of the joint project and Woomera itself although some sounding rocket flights did continue into the future. Recent Australian Activities: During the period in which ELDO was re-examining its operations including its launch site options, the Australian Department of Supply drew up a proposal to establish a launch site for ELDO near Darwin [Ref 2.56] in northern Australia. Ultimately this site, amongst a list of a dozen or so possibilities ended up as number two to the French proposal for a site at their launch base near Kourou in French Guiana. Australia’s part in space launch activities thus came to an end and there it stayed for many years. During the early 1980, possibly as a result of the acquisition of an Australian communications spacecraft, AUSSAT, engineering and public interest in space activities in Australia revived a little. In 1986 while manager of space systems at British Aerospace Australia, this author revisited the launch site problem and proposed an international launch site on Cape York Peninsula (CYIS7) to be used on a commercial basis by all nations needing a near-equatorial launch site. It was proposed that Australia was to provide all the ground facilities and each nation their own at the site. In collaboration with Mr. Stanley Schaeztel8, the then Premier of Queensland Sir Joh Bjelke-Peterson was contacted about the proposal. Sir Joh liked the idea and commissioned a study by the Institution of Engineers Australia. The study produced a feasibility study for the project [Ref 2.57] and a further scoping study was then commissioned by the Queensland Government Premier’s Department. This author was appointed space technology consultant to the Queensland government during the scoping studies which produced the documents of References 2.58-2.62. These separate studies looked at all aspects of the proposed launch site in detail and concluded that the project was viable. Although there was considerable international interest in the proposal it met with disdain from the Australian Government. The Australian Prime Minister at the time, the Hon. Paul Keating asked “What has space ever done for Australia?” As late as 1998 scholars were writing “…Aerospace development in countries like Australia face(s) obstacles in a political system not developed to encourage technological innovation and address associated risk” [Ref 2.63]. To this date it appears that nothing

6 ESRO: European Space Research Organisation 7 CYIS: Cape York International Spaceport 8 The then technical director of Hawker de Havilland in Australia

13

Chapter 2 Literature Review much has changed. A number of references are available at the location of ref 2.64 asking “Creating a Commercial Launch Industry: What will it take for Australia to Succeed?” During the scoping studies the Queensland Government was concerned about the lack of a firm commitment of a launch vehicle to launch from the proposed launch site. Consequently in 1987 a consortium was set up to propose a solid fuel launch vehicle to launch small scientific payloads up to 1 tonne mass into low earth orbit. The vehicle, Capricorn [Ref 2.65], was to be designed and built in Australia. The proposal was presented to the Queensland Government but no government or industry interest in the project was forthcoming so the project foundered. In 1989 Wilson studied the design of a launch vehicle intended for manufacture and operation in Australia utilising LOX/RP-1 propellant equivalent to Ariane 44L [Ref 2.66]. He discussed the justification for a national launch vehicle, the commercial aspects and its feasibility. He concluded that it would be technologically feasible to design and manufacture a launch vehicle in Australia, provided Government support was forthcoming and significant progress was made in the development of the local space industry. Additionally he felt that such a vehicle would be an important national asset. The work of the Capricorn consortium was repeated by the Southern Launch Vehicle consortium in 1992 [Refs 2.67-2.70] but by this time Orbital Sciences Corporation in the US had developed the Minotaur launch vehicle to fill the niche and the opportunity was lost. At least one advantage came from the spaceport and launch vehicle studies: After a Senate enquiry, in 1998 the Australian Government passed enabling legislation for space launch activities within the country [Ref 2.71]. A space licensing office (SLASO) was set up to handle the licensing of space launch operators. However in the typical fashion of the Australian government no actual action was taken to assist the development of a launch industry. An overview of opinion from space lawyers about the legislation can be found at Ref 2.72. In the early 1990s a group of amateurs set up a limited liability research institute (ASRI9) to develop the AUSROC series of light launch vehicles [Ref 2.73-2.75] and funding was finally 10 obtained by the CRCSS to construct FEDSAT [Ref 2.76], Australia’s third spacecraft, (after WRESAT and AO511 [Ref 2.77]). After the CYIS study a number of other space port proposals were advanced. The most advanced was by Asia Pacific Space Centre Pty Ltd. who proposed a launch site on Christmas Island using Aurora, a modified Soyuz launch vehicle [Ref 2.78]. The project gained a promise of financial support from the Australian Government, but not for the spaceport, only for common infrastructure such as port facilities. It wasn’t the Australian proposal that motivated Canberra but the Russian launch vehicles and the agreements with Moscow. Once again Australian capabilities and proposals were disparaged and the international capabilities were the most influential. Unfortunately finance for the space centre could not be obtained and once again the project foundered. In 2001 while at the US Air Force Institute of Technology, Rogers submitted a thesis [Ref 2.78] exploring the possibility of establishing an Australian indigenous space launch capability through developing and examining an Australian space launch program model.

9 ASRI: Australian Space Research Institute 10 CRCSS: Cooperative Research Centre for Satellite Systems 11 AO5: Australis-Oscar 5 – An amateur radio spacecraft constructed by Melbourne University Astronautical Society and launched as a secondary payload with Tiros by a US Delta Launch Vehicle in 1970

14

Chapter 2 Literature Review

Rogers based his model around the design of a solid propellant launch vehicle for scientific payloads. The model was statistically based, comparing Australian launch site location, vehicle design, program duration, and the percentage Australian indigenous input into the program with international data in order to obtain a probability for success of an Australian project. Rogers optimised the model in an effort to maximise the benefits of such a capability, namely political prestige, security and in-country technological base, while minimising the program's overall cost. He also concluded that it is within Australia’s capability to develop a space launch vehicle. In 2008 the Australian Senate Economics Committee held an inquiry into the state of Australia's space science and industry sector [Ref 2.79]. One of the most important problems identified by the committee was that Australia lacked both a space agency and a national space policy in order to formulate a whole-of-government approach to space issues. Gilbert [Ref 2.80] discusses Australia's historical political deficiency in relation to space before considering how space issues fit within broader policy frameworks at both the domestic and international levels. Using some of the findings from the Senate inquiry Gilbert addresses some core issues that the Australian government should consider when constructing a national space policy. However, despite the deficiencies identified by the inquiry the Government still refuses to form a space agency and has only [Refs 2.81, 2.82] appointed a space council to advise the Government. Although $40 million dollars has been made available in grants to space industries, the individual grants are insufficient to fund major projects. More successful Australian enterprises are the hypersonics projects originating at the University of Queensland [Ref 2.84]. The HyShot flight program is an experiment designed to develop a correlation between pressure measurements made of supersonic combustion in the University of Queensland's T4 shock tunnel, and that which is observed in flight. Supersonic combustion was achieved in the HyShot II and III flights. HyCAUSE (Hypersonic Collaborative Australia/United States Experiment) is a DARPA- sponsored, scramjet technology program involving researchers in Australia and the United States [Ref 2.85]. The program began in April 2004 and consisted of ground tests and CFD analyses of two-dimensional, inward-turning scramjet engine concepts. A flight vehicle incorporating an inward-turning scramjet engine model was developed with a flight test aimed at collecting engine data at Mach 10 flight conditions. In November 2006 DSTO signed a $74 million Hypersonics International Flight Research Experimentation (HiFire) Agreement with the [Ref 2.86]. Up to ten hypersonic flight experiments are planned to occur at Woomera over five years under the agreement. In Round 1 of its “Australian Space Research Program” the Australian government, under the Auspices of the newly formed space policy unit, has recently funded the Scramjet- Based Access-To-Space Systems (Scramspace) project. Scramspace is the first phase of a stepping-stone-based roadmap to develop a scramjet-based access-to-space industry. A scramjet (supersonic combustion ramjet) is an air-breathing combustion engine which can be combined with rockets to produce a more fuel-efficient hybrid launch system for access to space. This project aims to answer key scientific and technological questions through both flight and ground tests, leveraging Australia’s world leadership in scramjet research and development. It will contribute to Australia’s assured and secure access to space and cutting-edge space technology. Importantly, the project is intended to build an industry- ready talent pool for a future Australian scramjet-based access-to-space industry.

15

Chapter 2 Literature Review

French Launch Vehicles: On 12 June 1945, only weeks after the end of the Second World War, the newly reconstituted French Ministry of War created the Centre for Study of Guided Missiles with the aim of acquiring for German V-2 rocket technology developed during the war. An initial group of over thirty German engineers was recruited for this purpose and established the Laboratory for Ballistic and Aerodynamic Research (LRBA) at Vernon. In August 1946 this group sketched out the development steps that would eventually lead to the Ariane rocket of the 1980's. First the necessary production base and test facilities for French production of the Super-V-2 using nitric acid and kerosene propellants would be created in France. However by early 1947 problems arose in that the American and Russian 'allies' would not co- operate in providing key V-2 components. Therefore it was decided to proceed on a dual track, still with the objective of beginning flight tests in 1952. The German team at LRBA assisted in French development and flight test of the A9 rocket, picking up from where the design had been at Peenemünde in 1944, under the project 4211. A second team under Jean-Jacques Barre pursued the development of his 'pure French' 1941 rocket design, under the project 4212. But there was no government interest in providing the substantial funding necessary to bring a prototype to flight status. The Super V-2 was cancelled in 1948 and the team concentrated on development of the one-tenth scale Project 4213/Veronique with a 4 tonne thrust motor. However the Super V-2 studies had shown the way. Veronique began flights in 1950, beating the 'all-French' Eole by two years. Veronique set the course for rocket development and propellant selection in France that would finally lead to the Ariane space booster twenty years later. Michels and Przybilski [Ref 2.87] give a history of the Super V-2 in their book Peenemünde and its inheritance in East and West. Development of the Veronique began in March 1949, with the primary objective of providing a flight test vehicle for liquid rocket engine development and for scientific flights. Combustion instability held up development and after a nine-year program the much larger Veronique-61 sounding rocket emerged as the final version. Wade gives descriptions of the French vehicles in his Encyclopedia Astronautica [Ref 2.88]. [Ref 2.88] was a step toward larger liquid propellant launch vehicles, building on the Veronique experience. It burned 12.8 tonnes of nitric acid/turpentine pressure-fed propellants in 91 seconds. The engine was gimbaled for pitch and yaw control, with roll controlled by aerodynamic fins. Propellant sloshing caused the first three launches to fail. This was remedied in the later tests. Variants of the two-stage Saphir (Saphire) vehicle were designed to allow testing of radio- controlled guidance, inertial guidance, warhead separation and re-entry of an ablative reentry vehicle. Addition of a third stage would transform Saphir into the () space launcher [Ref 2.55, pg 58]. Diamant-A launched the Asterix spacecraft and was followed by the Diamant-B launcher. Ariane Family: In 1966 the US controlled Organisation (later renamed Comsat) had the internationally agreed responsibility for international satellite communications. When France and Germany wished to have a joint communications spacecraft launched () the US refused to launch it on the grounds that it would be financially prejudicial to Intelsat and a threat to the monopoly of the American launchers and spacecraft makers. Following the Symphonie Saga and the final failure of the ELDO Europa launch vehicle France became convinced that Europe needed its own launch vehicle and proposed a new European launch vehicle called L3S. At the ELDO meeting of 1972 ELDO and ESRO

16

Chapter 2 Literature Review were merged and the members agreed to build the L3S launcher. L3S was specifically designed to launch spacecraft to geosynchronous orbit. Subsequently, the European Space Agency (ESA) was formed on 31 July 1973 and funding and organisational arrangements were settled. Harvey [Ref 2.55, Chap. 4] gives an interesting outline of the process through which ESA and L3S were agreed. L3S became the Ariane launcher the first launch of which was Christmas Eve 1979. The launcher evolved through four models of increasing payload capabilities which evolved through to Ariane 4. Reference 2.89 gives a summary of the Ariane program and statistics of the successes and failures of the launches from the first AR1 to AR4, while Harvey [Ref 2.55, Chap. 4] gives a forty page summary of the development of the Ariane system through to and the proposed 15 tonne Ariane-5. References 2.90 and 2.91 are the user manuals for the Ariane 4 and Ariane 5 launchers and give a perspective from the user’s point of view. They outline the spacecraft checkout facilities, operations and planning, safety and launch constraints, documentation requirements for use of the system and the interfaces between the spacecraft and the launcher. Furniss [Ref 2.92] discusses the third stage problems that occurred early in the flight program and the fixes to the HM-7 third stage engine. In reference 2.93 he goes on to summarise the design and operation of Ariane-4 in The Big Shot. Dulout discusses the performance of the launch vehicle in terms of the required orbit and the injection accuracy and then goes on to discuss the accuracy of orienting the spacecraft before separation from the launcher. Breton [Ref 2.95] discusses the launcher’s capabilities into orbits other than geostationary while Heydon and Weinrich discuss the vehicle’s performance through the ‘90s and beyond [Ref 2.96] referring to possible developments envisaged at the time. The development program for Ariane was so tight in time that the decision was taken to use standard materials rather than attempting to develop advanced materials. Desloire discusses the materials choices and the problems associated with the choices [Ref 2.97]. Essentially the materials chosen were various grades of Aluminium, Stainless steel and Carbon fibre. The success of Ariane relies on the high accuracy with which it can place a spacecraft into the desired orbit, its ability to launch multiple spacecraft in one flight and the accuracy with which they can be pointed at release. Kemp (GEC Ferranti) gives a complete overview of the Ariane Inertial Measurement Unit (IMU) which evolved from the British Black Arrow IMU. He also describes the overall guidance and control system of the launcher in Steering Ariane [Ref 2.98]. Dulout and Spagnulo, [Ref 2.99] describe the Vehicle System for In-Orbit Payload Separation and Waterman describes the Guidance Law for the Ariane On-Board Computer [Ref 2.100], Angus et al describe The Guidance System for Ariane in the Journal of Navigation [Ref 2.101] and Cutcher and Smith describe the Onboard Flight Control in a JBIS article [Ref 2.102]. The first and second stages of the Ariane 1-4 vehicle propulsion based around UDMH/N2O4 storable propellant technology of the engine is described by Souchier et al in The Development of the Viking Engine [Ref 2.103]. Pouliquen and Gill describe the cryogenic technology of the third stage engine in Performance Characteristics of the HM7 Rocket Engine for the Ariane Launcher [Ref 2.104] while Faure describes the development of the engine including the difficulties such as hard starts and its fixes [Ref 2.105]. Moving forward to the Ariane 5 vehicle Guillard and de Boisheraud describe the Ariane 5 H155 Main stage in the ESA Bulletin [Ref 2.106], Holger Holsten of MBB/ERNO describes the development of the Ariane 5 upper stage [Ref 2.107]. Gary Hudson of Pan American Launch systems described his experience designing the pressure fed

17

Chapter 2 Literature Review launch vehicle as “We spent far more money designing the pressurisation system than we did designing the main engine” [Ref 2.108]. Dussollier and Teissier describe the Ariane 5 Main Stage Oxygen Tank Pressurization. As an example of the trade-off optimisations possible, the oxygen tank is pressurised by helium stored as a liquid which saves 400kg weight rather than if it was stored as a high pressure gas [Ref 2.109]. Looking forward to future improvements of the AR5 launcher, a trade-off study replacing the solid propellant strap-on boosters with liquid propellant boosters comparing LOX/LH2, LOX/Kerosene and LOX/Methane propellants is described by Vernin and Pempie [Ref 2.110] and in more detail by Burkhardt et al in reference 2.111. Interestingly it was shown that LOX/Methane gave a weight and cost disadvantage at the system level especially if one could draw on heritage LOX/Kerosene experience. However in a reusable booster the lower operational and maintenance costs gave LOX/Methane an advantage.

2.7 JAPAN, CHINA, INDIA.

2.7.1 Japan Japan became the fourth nation after France to launch an earth orbiting spacecraft using its own launch vehicle. L Series: Japan's first family of orbital launch vehicles was the , or L series. These rockets were solid propellant and were scaled-up versions of the Kappa sounding rockets. The fifth launch of an L-4S placed Japan's first artificial satellite, , in orbit on 11th February 1970. The spacecraft re-entered the atmosphere on 2nd August 2003. The spacecraft has its own web page on the JAXA12 web site [Ref 2.112] and the L-4S launcher at [Ref 2.113]. M Series: The L series was followed by the first generation of the series. The four-stage M-4S rocket utilised a gravity turn for orbital insertion and was fin and spin stabilised. It could place a 180kg satellite into low earth orbit [Ref 2.114]. The second generation three- stage M-3C was equipped with secondary injection thrust vector control and side jet systems to improve the accuracy of orbital injection and could place 195kg into low earth orbit. The M-3H increased propellant to achieve a greater payload capability of 300kg by extending the first stage motor casing. The third generation of the Mu series was M-3S rocket which could also place a 300 kg satellite into low earth orbit. With new upper stages and an optional fourth stage the M-3SII series was the fourth generation of the Mu series capable of 770kg into low earth orbit. The M-V series was a technology development of the M series capable of launching 1800kg into LEO and 500kg on interplanetary missions. It was regarded as the best solid propellant launcher in the world. First launched in 1997, its production was unfortunately discontinued in 2003. The excellent article by Morita describes the development and use of the M-V launcher [Ref 2.115]. N series: This series of rockets were license built versions of the US Delta rocket built in Japan using American and Japanese components and used to launch NASDA13 spacecraft. The N-1 could place up to 130kg into GSO. In the period between 1975 and 1982 seven of these vehicles flew of which six were successful. The larger N-2 successfully launched eight spacecraft in eight launches in 1980-86. The N-2 had a GSO payload capacity of 350kg. Kyle has written a web article describing the history of the Thor missile with a page on the Japanese “Deltas” [Ref2.116]

12 JAXA: Japanese Aerospace Exploration Agency 13 NASDA: (Japanese) National Space Development Agency (now JAXA)

18

Chapter 2 Literature Review

H series: The H series launchers in order of development were the H-1, H-2 and H-2A (and variants). The H-1 was a license-built version of the US Delta rocket, whereas the H-2, H-2A, and H-2B are entirely Japanese in design and construction H-1: The first stage of the H-1 was essentially the same as that of the N-2 with a LOX/Kerosene main engine and six to nine small solid-propellant strap-on boosters. The second stage, built by Mitsubishi Heavy Industries, was of Japanese origin and burned LOX/Liquid Hydrogen propellant. The third stage, designed by Nissan, was solid propellant and enabled payloads of up to 1,100kg to be placed into GTO. The H-1 program operated between 1986 concluded in 1992 with nine successes and no failures. Reference 2.42 by Isakowitz et al gives details of the M-V, J-1, and H-II vehicles but not the H-1or the H-IIB. H-2: As the Delta licensing agreement restricted the use of the H-1 for commercial flights the H-2 was developed based on all-Japanese propulsion systems. It was designed to provide greater payload capacity and to permit unrestricted commercial space transportation. The H-2, which made seven successful flights between 1994 and 1999, could lift payloads four times heavier than those of the H-1, up to ten tons into LEO or up to four tons into GTO. It opened the door for JAXA interplanetary spacecraft designed to explore the moon and planets. Considerably larger than its predecessor, the H-2 consisted of a two-stage core vehicle, burning LOX and liquid hydrogen propellant in both stages, with two large four segment solid-propellant strap-on boosters designed by Nissan. Wade [Ref 2.117] gives details of all the H-2 launch vehicles through to 23rd February 2008. H-2A: An upgraded version of the H-2 currently in service; the first H-2A was launched successfully from Launch Pad 1 at Tanegashima on 29th Aaugust 2001. Intended to compete commercially on the world market, the H-2A builds upon its predecessor and incorporates a simplified design and upgraded avionics and engines. Although the core vehicle is similar to the H-2's, the H-2A uses new solid and liquid boosters to improve payload performance. On November 28th 2009 H-2A number F16, lifted off from Tanegashima Pad 1 and twenty minutes later orbited the Optical 3 Information Gathering Satellite (IGS) into a polar orbit. It was the 10th consecutive H-2A launch success, and the 15th success in 16 flights. JAXA web site details the launcher at reference 2.118 H-2B: The H-2B takes advantage of parts proven by the less powerful H-2A. It uses a larger, "widebody" first stage with a diameter of about 5.2 metres housing larger propellant tanks that feed dual hydrogen-fueled LE-7A main engines. The H-2B's second stage and solid rocket boosters are unchanged from the designs used by the H-2A rocket. JAXA successfully launched the first H-2B carrying an unmanned cargo module from Tanegashima early on Friday 11th September 2009. The cargo module, called the H-2 Transfer Vehicle (HTV), will transport supplies to the International Space. JAXA has provided a special web page site to describe the launch vehicle, the HTV and the mission [Ref 2.119].

2.7.2 China The development of the Chinese rocketry and space technology was led by American-trained Tsien Hsue-Shen. Tsien arrived in the United States on a Boxer scholarship in 1935 and became a protégé of Theodore von Karman. In 1950, his security clearance was revoked, ending his ability to conduct further research. In September 1955 he returned to China, By 1970 Tsien had launched China's first satellite, the DFH-1, using his CZ-1 (Chang Zheng14-1) rocket (a DF-2 missile with an upper stage), making China the fifth spacefaring country in the world. The improved CZ-2 was developed into an extensive

14 Chang Zheng = Long March

19

Chapter 2 Literature Review launch vehicle family over the next thirty years. It was used for launches of the FSW photo , with a recoverable re-entry capsule beginning in 1974. The CZ or Long March family of launchers has been developed to become China's primary family. The Shenzhou manned spacecraft was launched by Long March in 2003 and the Chang'e 1 lunar orbiter in 2007. The maximum Long March payload to LEO is 12,000 kg (CZ-3B) and to GTO is 5,500 kg (CZ-3B/E). (LM-5, CZ-5, or Changzheng 5) is the next generation Chinese heavy lift launch system currently under development by China Academy of Launch Vehicle Technology (CALT) and due for its first launch in 2014. Six CZ-5 vehicle configurations are planned for different missions, with a maximum payload capacity of 25,000 kg to LEO and 14,000 kg to GTO. The Long March 5 will have the second largest payload of any rocket after Boeing's Delta IV Heavy. References detailing the are: Wade [Ref 2.120], who gives a history of the program and the Wikipedia web resource that gives details of the Long March family [Ref 2.121]. Isakowitz [Ref 2.42] also has a chapter devoted to the Long March launcher.

2.7.3 India Gupta, Suresh and Sivan [Ref 2.122] state: “[A] launch vehicle is a critical element in the self-reliant programme of space endeavour, because of the vicissitudes of geopolitics and non-availability of the know-how from those who possess this technology, because of the various dual use based control regimes, particularly, [the] Missile Technology Control Regime. Considering that access to technologies, components, materials etc. is under stringent technology control regimes of the developed countries, all-round indigenous effort by ISRO, in association with national R&D institutions, academia and industry to develop the complete range of technologies was called for [for] the development of launch vehicles.” This belief led ISRO15 to develop two major launch vehicles to support the Indian space exploration effort. Experience was first gained through the 1960s to 70s by developing technologies for the use on sounding rockets and an experimental phase in the 1980s developing the SLV-3 and the Augmented Satellite Launch Vehicle (ASLV). The four stage SLV-3 could launch 40kg into LEO while the ASLV which was an SLV-3 with two strap on solid boosters could launch 150kg into LEO. During this time the design and manufacturing infrastructure was also created. As a result of this experience ISRO undertook the development of the PSLV (Polar Space Launch Vehicle) and the GSLV (Geosynchronous Space Launch Vehicle). The PSLV had an interesting architecture in that while the first stage was solid propellant with two solid propellant strap on boosters, the second stage was earth storable liquid propellant, the third stage was solid and the fourth stage was again liquid propellant. New technologies (for India) were introduced on the PSLV. These included gimbaled nozzle control and a REdundant Strap-down Inertial Navigation System (RESINS). PSLV has a capability of placing up to 1900kg into Sun Synchronous Polar Orbit and 600kg into LEO. GSLV was developed concurrently with PSLV. The design mission for GSLV was to place a two tonne class spacecraft into Geosynchronous Transfer Orbit (GTO). It is a three stage vehicle with a solid propellant first stage core and four N2O4/UDMH strap-on boosters. The second stage is also liquid propellant while the third stage is a LOX/LH2 cryogenic stage.

15 ISRO: Indian Space Research Organisation

20

Chapter 2 Literature Review

With the addition of a solid propellant (PAM) GSLV is also capable of placing navigation spacecraft directly into 19000km circular orbits. Following the success of GSLV the advanced GSLV Mk-III vehicle development was started with the aim of launching a 4 tonne class spacecraft into GTO. The architecture is that of a liquid first stage core with two strap-on solid boosters and a cryogenic LOX/LH2 upper stage. Associated with the launch vehicle, development of a range of improved technologies and development and operational facilities was commenced. In addition, air breathing propulsion and reusable launch vehicles and manned space flight projects are under way. The 18 page article by Gupta, Suresh and Sivan is an excellent overview of the Indian launch vehicle program. It contains a number of colour illustrations of the various Indian launch vehicles and ground facilities. A second article [Ref 2.123] by Gupta and Suresh gives a similarly detailed overview of the Development of Navigation Guidance and Control Technology for Indian Launch Vehicles. Wade [Ref 1.124] also has a page on the Indian space program. Isakowitz et al [Ref 2.42] include a chapter on description, performance, operation, production and history of the PSLV and ASLV.

2.8 SCUD SS-1 Scud is the NATO reporting name for a series of tactical ballistic missiles developed by the during the and widely exported to other countries. The Russian names for the missile are the R-11 (the first version), R-17 (Scud-B) and R-300 Elbrus (Scud-C). There are a large number of variants developed by other countries. The SCUDs are versions of the German World War II V-2 missile modified to use storable propellants Kerosene and Nitric Acid. They are still steered by graphite vanes in the exhaust as was the V-2. The al-Hussein is an Iraqi modification of the R-17 that doubled the missile's range at the expense of more than halving the payload and accuracy. launched 331 Scud-B missiles at during the Iran-Iraq war and 189 al-Hussein missiles at Iranian cities during the 1988 "War of the Cities." In the first Iraq War, between January 18 and February 26, 1991, 40 Scuds were launched by Iraq against and 46 against . This handful of launches had enormous operational impact on coalition operations requiring the diversion of resources to counter possible effective attacks by the missiles [Ref 2.8]. Wade discusses the Scud in his Encyclopedia Astronautica [Ref 2.125] while references 2.126 and 2.127 give some more detailed figures and operational details. In a CRS16 report to Congress, Feickert discusses Iran’s ballistic missile capabilities, mainly Scud-Bs but also their extended range Shahab-3 missiles [Ref 2.128]. The R-17 VTO (SS-1e Scud-D) was an attempt to enhance the accuracy of the R-17. However, by the time it was ready more advanced weapons were in use and the Scud-D was not acquired by the Soviet armed forces. Instead it was exported as an upgrade for Scud-B users in the 1990s. Unlike previous Scud versions, the Scud-D had a warhead fitted with its own terminal guidance system that separated from the missile's body. With a TV camera fitted in the nose, the system could compare the target area with data from an onboard computer map library possibly attaining a Circular Error Probability (CEP) of 50 m, while retaining the 300 km range of the Scud-B [Ref 2.129]. Pinkston [Ref 2.130] discusses long range ballistic missile aspirations with its Hwasŏng-6. In designing the Hwasŏng-6 North Korean missile engineers could have

16 CRS: Congressional Research Service

21

Chapter 2 Literature Review benefited from wreckage of Iraqi al-Hussein missiles provided by Iran. The al-Hussein was a modified Scud-B with a range of about 600km. Iraqi engineers were to double the range of the Scud-B by extending the oxidiser tank by 0.85 meters and the fuel tank by 0.45 meters, and by reducing the mass of the warhead from 1,000kg to 500kg. The Hwasŏng-6 has a range of 500km with a warhead of 770kg, and is reportedly more accurate than the al-Hussein.

2.9 LAUNCH VEHICLE DESIGN Propulsion: One of the earliest detailed texts on rocket propulsion is the extensive work by Barrère et.al [Ref 2.131]. The authors’ text stems from a series of lectures at the Université Libre de Bruxelles given in 1955 and covers the theoretical aspects of rocket propulsion. Chapters cover: Rocket propulsion elements, Nozzle theory and characteristic Parameters, Aerothermochemistry of combustion products, Solid propellants, Internal ballistics, Design of solid-propellant rockets, Liquid propellant rockets, Experimental techniques for rocket propulsion, Liquid propellants, Combustion instability, Rocket performance and Variational methods in optimising rocket performance. Overall this is an extremely detailed theoretical text. In 1967 NASA published its monograph NASA SP-125, Design of Liquid Propellant Rocket Engines [Ref 2.132]. This is a workbook and text dealing with the design of both pump-fed and pressure-fed liquid propellant rocket engines (LPREs). The reader is led through the gas dynamics of the rocket engine and using worked examples a number of engines are sized along with their turbopumps or pressure feed equipment. This is an excellent source for a beginner to get a basic understanding of the calculations involved with the design of LPREs. A further updated edition of the work was published by the AIAA in 1992 [Ref 2.133]. One of the major problems occurring during the development of a rocket engine is that of combustion instability. The first important book published on the matter was Crocco and Cheng, Theory of Combustion Instability in Liquid Propellant Rocket Motors [Ref 2.134]. In 1974 NASA published its SP-194, Liquid Propellant Rocket Combustion Instability, edited by Harrje and Reardon [Ref 2.135]. This monograph was contributed to by many authors from many different organisations and took several years to compile due to the difficulty of obtaining consensus as to what were the important aspects of the subject. Over a period of about forty years G.P. Sutton has been the author of an important work on Rocket Propulsion Elements [Ref 2.136] that has seen many editions. At the time of writing the eighth edition has been released. Along with his History of the Liquid Propulsion Rocket Engine [Ref 2.25] it remains one of the standard texts on the subject. Cornelisse et al authored Rocket Propulsion and Spaceflight Dynamics published by Pitman [Ref 2.137]. The book includes chapters on Celestial mechanics, Dynamics, Chemical rocket motors, both solid and liquid and launch vehicle and spacecraft motion. This is an intermediate level, fairly theoretical, text. Of more practical use, again partly a workbook is the text by Humble, Henry and Larson [Ref 2.138] Space Propulsion Analysis and Design. The text works through the process of designing a propulsion system from initial mission definition, choice of propulsion type to the sizing of the propulsion subsystem. This is a useful practical book. Finally two texts present historical case studies of the design and development process and the problems both engineering and political that can occur during the development of LPREs. The books are Rocketdyne: Powering Humans into Space [Ref 2.139] and Taming Liquid Hydrogen: The Centaur Upper Stage Rocket, 1958-2002 [Ref 2.140]

22

Chapter 2 Literature Review

Engineering: The Handbook of Astronautical Engineering edited by Koelle [Ref 2.141] is more of an overall engineering perspective of the whole of astronautical engineering. It is a monumental work by many authors but, with the developments in electronics since its publication, part of the text is now out of date. Much of the work is still useful A more recent text based on the continuing lecture series at the University of Southampton, UK is Spacecraft Systems Engineering [Ref 2.142]. The text examines each subsystem constituting a spacecraft: Mechanical, Electrical, Propulsion, Thermal, AOCS, launch and operations. The spacecraft is built up in a systems approach using the fundamental building blocks of the sub-systems. In this edition a chapter has been added relating to the systems engineering of small spacecraft. An immense work distilling the experience of the engineers involved in NASA projects is the NASA Special Publication Series SP-8001 to SP-8126, Design Criteria [Ref 2.143]. These documents cover Chemical Propulsion, Structures and other aspects of spacecraft design. They layout the state of the art of the design of the item in question followed by the criteria to be met and requirements to be specified in the design. As such they are an essential aid for the spacecraft designer. The series includes many documents on LPRE design including combustion chambers and turbopumps. Hammond has written two texts in the AIAA Education Series [Refs 2.144, 2.145] dealing with the design of space transportation systems which are more specific to launch vehicles than Ref 2.142. The first text by Hammond deals with systems engineering and systems design followed by systems architecture and then examines the US as a reusable system. Hammond then examines expendable LVs and reusable LVs in general. A chapter is devoted to each of: Operations and support, Cost analysis, System optimisation, Development and Management. Hammond also gives an extended bibliography for each chapter including a list of web resources of some quite technical subjects. This text is a good readable basic introduction to system design of space transportation systems. Hammond’s second text is more detailed than the first and concentrates more on subsystem approach than his more general first text. Chapters deal with: System engineering and design overview, Conceptual design and tradeoffs, Design Sequence; Aerothermodynamics; Thermal heating; Structures and materials; Propulsion; Flight mechanics and trajectories; Avionics, Computers and control systems; Multidisciplinary optimisation; Human factors; Payloads and integration; Launch and mission operations; Safety and mission assurance. In addition there is an essential reading chapter on “Some Space Transportation Systems Lessons Learned” as well as a CD of design software. These two texts are essential reading for a launch vehicle designer; however they come across as being influenced by the current United States thinking regarding reusable launch vehicles. It is this author’s opinion that in the Australian situation the simplest possible expendable launch vehicle should be the objective so that some of the drivers of the thinking by Hammond may need to be modified for an Australian ELV project. Propellants: Edwards [Ref 2.146] reviews the historical development of fuels and propellants for aerospace propulsion. For rocket propulsion Edwards examines the classical propellants such as Kerosene, RP-1, and the Acids RFNA and IRFNA and discusses their merits and drawbacks. He goes on to discuss storable and cryogenic propellants and future trends in fuels and propellants. Information is included on the various difficulties encountered and the solutions to the use of various propellant combinations. The review includes 173 references.

23

Chapter 2 Literature Review

The current trends in rocket propellants are away from the use of the storable NTO/Hydrazine(s) due to the poisonous nature of the components and exhaust products. In an Aerospace America discussion article Sietzen [Ref 2.147] outlines the growing concerns over the use of toxic and polluting propellants. This is particularly true of the exhaust products of solid propellant boosters such as those used as strap-ons by the US Space Shuttle and the Ariane-5 vehicles which produce concentrated hydrochloric acid as an exhaust product. Sietzen discusses possible solutions but concludes that less toxic propellant combinations with benign exhaust products are still in the future. A more technical discussion of the same problem examining the use of alternative propellants with performance figures for various missions included is given by Valentian et al [Ref 2.148]. They include figures for toxicity and handling problems as well as mission trade-offs. As an example of what is probably the highest performing green propellant Palaszewski [Ref 2.149] discusses gelled liquid Hydrogen including 5% Methane gelled in LH2 which gives approximately 4sec greater Isp with LOX than LH2 alone. Density is also increased. The performance figures of gelled hydrogen and other propellant combinations can be calculated using the CEA computer program. This program (Chemical Equilibrium with Applications) is a result of the long-term work carried out by the NASA Glenn (was Lewis) Research Centre who have carried out investigations into the theoretical performance of propellants. The theoretical methods of carrying out the calculations include minimisation of the Gibbs free energy and are described in references 2.150 and 2.151. Pratt and Whitney provide a poster [Ref 2.152] that sets out the results of such calculations for many propellant combinations at optimum expansion to sea level (101.325 kPa, 14.7 psi) and at ε =40 in vacuum. Orbital Mechanics & Operations: A basic text covering most of the aspects of orbital mechanics is that by Curtis [Ref 2.153]. He covers basic particle dynamics, the two body problem, the orbit in three dimensions and preliminary orbit determination as well as orbital manoeuvres. Also covered are rendezvous and interplanetary trajectories. Rocket vehicle dynamics including optimal staging comprise the last chapter. More technical but covering similar material is the standard text by Roy, Orbital Motion, [Ref 2.154]. Bate, Mueller and White present a text on Astrodynamics that is based on the lectures given to the US Air Force Academy [Ref 2.155]. They concentrate heavily on orbit determination, an important problem for determining the trajectories of incoming ballistic missiles. In Introduction to Space Dynamics, Thomson [Ref 2.156] concentrates on rigid body dynamics but provides a very relevant (to this thesis) chapter on performance and optimisation of launch vehicles including stage and flight trajectory optimisation including propellant utilisation. He describes the gravity turn mathematically. A well known standard text on mission planning is Space Mission Analysis and Design (SMAD) [Ref 2.157] but more relevant for the baseline mission of most launch vehicles is the text by Pocha [Ref 2.158]. This text originated as the handbook for mission analysts at British Aerospace space and communications division at Stevenage UK and concentrates on mission planning specifically for geostationary launches and primarily for the Ariane launch system. Launch, transfer orbit, the apogee manoeuvre, drift orbit and station keeping are covered, each by separate chapters. The on-station phase which comprises most of the lifetime of the spacecraft is covered in chapters on spacecraft operations, orbit propagation and tracking and orbit determination. Spacecraft stability is discussed in the final chapter.

24

Chapter 2 Literature Review

More detailed and concentrating on station keeping is the text by Soop [Ref 2.159]. This text originated as an ESA publication in 1983. It was originally the theoretical background for orbit control of geostationary spacecraft by means of the PEPSOC software package at the European Space Operations Centre (ESOC) at Darmstadt, Germany. Despite a number of decades and many hundreds of space launches and although much effort has been made to reduce it, the cost of a launch is still very high. Wertz and Larson [Ref 2.160] have edited a text that concentrates on the reduction of space mission costs. They show that the cost of a launch vehicle is 97% labour and 3% materials. It is also shown that for the lifecycle of a launch vehicle system 60% of the cost occurs in the operational phase rather than in the development phase. This is an important factor for reducing launch costs: if operational costs can be reduced, e.g. less staff, then launch costs can be reduced. This text will be important for the planning of any Australian space launch system.

2.10 NUMERICAL METHODS A useful text for some general methods of data reduction and error analysis is that by Bevington [Ref 2.161]. Originally used for the nuclear sciences the text includes thirty routines in FORTRAN-II for such purposes as evaluating the Chi squared test, fitting to straight lines and polynomials as well as multiple regression, interpolation, integration and smoothing. A much more comprehensive text is that by Press et al [Ref 2.162] Numerical Recipes in Fortran 77 and its companion text Numerical Recipes in Fortran 90 [Ref 2.163]. Of particular interest to this thesis are the chapters on sorting (Chapter 8) and minimisation and maximisation of functions (Chapter 10). For sorting the authors give a useful routine to carry out a bubble sort and for minimisation they give a routine for carrying out the Nelder and Mead Downhill Simplex Method in Multidimensions. The original paper on this method is given in reference 2.164. If one needs special functions such as Bessel functions they are also included. A word of warning to the reader though: Not all of the routines may work well. This author’s own routine for calculating the Bessel function J0 works far more efficiently and accurately than that given by Press et al. If special functions are required such as the Cosine and Sine integrals17 reference 2.165 is useful as it is dedicated specifically to the Computation of Special Functions. Evolutionary algorithms are ideal candidates for solving multi-objective optimisation problems due to their population based approach. A text by Deb, Multi-Objective Optimization using Evolutionary Algorithms [Ref 2.166] discusses the principles of multi objective optimisation as well as a number of classical approaches. A second text by the same author [Ref 2.167] deals specifically with Optimization for Engineering Design. He discusses many minimisation methods and gives FORTRAN routines for a number of them. Of particular interest, he also covers the simplex method of Spendley (1962) which was modified by Nelder and Mead (1965) [Ref 2.164]. The major minimisation method used in this thesis is the evolutionary (genetic) algorithm. Deb covers this topic extensively and gives a software package for carrying out the method; however the evolutionary algorithm software utilised in this thesis is by Ray and Sarker. Reference 2.168 gives an example of the use of the software by these authors in Solution of Gas Lift Optimization Problems. The method is based on the NSGAII method of Deb et al [Ref 2.169]. Ray and Liew use a swarm metaphor for Multi-objective Design Optimisation [Ref 2.170] based on a concept of the behaviour of a real swarm in which the members of the swarm

17 Sometimes required for upper stage performance calculations

25

Chapter 2 Literature Review update their direction through communication with a leader in order to achieve a common goal. The behaviour of the stack optimisation population discussed in this thesis is similar to that of a swarm discussed by Ray and Liew. As examples of the use of Evolutionary algorithms in the aerospace field, Ross et al used GA18 to carry out A Fast Approach to Multi-Stage Launch Vehicle Trajectory Optimization [Ref 2.171], while Mondolini used A Genetic Algorithm for Determining Optimal Flight Trajectories [Ref 2.172]. Wollam et al discuss Reverse Engineering of Foreign Missiles via Genetic Algorithm [Ref 2.173] while Bailey has done just that by Reverse Engineering of a SCUD Missile using a Genetic Algorithm for his MS thesis [Ref 2.174]. Hartfield’s group at Auburn University, Alabama, has written extensively about reverse engineering of missiles using GA. Two of their papers Reverse Engineering of Solid Rocket Missiles with a Genetic Algorithm [Ref 2.175] and Genetic Algorithm Optimization of Liquid Propellant Missile Systems [Ref 2.176] were delivered at the AIAA 45th Aerospace Sciences Meeting and Exhibit in Reno, Nevada, 2007 along with this author’s paper Evolutionary Algorithm use in Optimisation of a Launch Vehicle Stack Model [Ref 2.177] where we were attempting to reverse engineer the Ariane 44L launch vehicle. Geethaikrishnan has carried out similar work in Multi-Disciplinary Design Optimization Strategy in Multi-Stage Launch Vehicle Conceptual Design [Ref 2.178] and in Geethaikrishnan et al, Genetic Algorithm Guided Gradient Search for Launch Vehicle Trajectory Optimization [Ref 2.179]. Chartres carried out optimisation of terminal area trajectories for reusable launch vehicles for his PhD thesis at the University of Adelaide in 2007 [Ref 2.180]. He utilised gradient methods, genetic algorithms and simulated annealing methods and included various errors in the ascent and terminal area descent such as navigation errors, steering command errors and initial condition errors to determine off-nominal flight parameters. Sobieszczanski-Sobienski and Haftka gave a survey of recent developments in multidisciplinary aerospace design optimisation [Ref 2.181] as they stood in 1997. Interestingly, they only mention genetic algorithms and simulated annealing as an appendix at the end of the paper under the heading “Design Space Search”.

18 GA: Genetic Algorithms

26 Chapter 3 System Considerations

3.1 SYSTEM BREAKDOWN While the launch vehicle itself is the most publicly visible component of a space launching system, other parts of the system are no less important. The development and construction of these other components can consume large amounts of the overall cost of the total system and can contribute to the system’s success or failure in a major way. They also impact on the design of the launch vehicle itself. Therefore, it is appropriate to regard these components, in the Australian context, as major considerations for any trade studies of systems.

Table 3.1: Launch vehicle system components and their major impacts/impactors Major Component Major Impact Sub-system Ground • Launch Pad • Location, Access, Transport Facilities • Assembly & Checkout Facilities • Turnaround time, Time on Pad • Trajectory • Range Safety • Guidance, Tracking and Safety • Orbit Accessibility • Recovery Facilities • Qualified Staff • Operations & Management • Education Launch Vehicle • Baseline Mission • Launch Vehicle Size • Structure • Materials • Propulsion • Propellant Selection • Thermal • Electronics Availability • Avionics & Telemetry • Trajectory • Guidance and Control • Telemetry/Tracking • Aerodynamics Manufacturing • Design • Capability • Manufacturing Facilities • Equipment Availability • Test & Calibration Facilities • On-going Improvement/Upgrading • Materials Choice • Materials Supply • Transport • Structure Size and Fragility Costs • Design & Proving Cost • Complexity of Design • Ongoing Manufacture Cost • Customer Satisfaction • Operations Cost • Operational Flexibility Politics • • Many

Table 3.1 above sets out one possible breakdown of the components required for a total launch vehicle system. The major impacts of the choice of components are also shown. Within the Australian scenario, almost all of the sub-system components would need to be developed in order to implement a launch vehicle system as little remains of the former infrastructure at Woomera. This is both good and bad. It is bad from the point of view of the cost and effort to rebuild. It is good considering that optimal sub-systems and components for an operational system could be designed and implemented. As most of the constraints on a system throughout its entire life-cycle are determined by decisions made at the back-of-the-envelope design stage, it is important that these first decisions are made correctly.

27

Chapter 3 System Considerations

For example, when the British Blue Streak had been cancelled as a weapon and its utility as a launch vehicle was being studied; an upper stage was being designed by Saunders-Roe [Ref 1.4]. A stage diameter of 54 inches was causing difficulties with control authority and a larger stage diameter was preferable. A diameter of 58 inches was the maximum possible due to facility constraints at the High-Down test site. Had the test facilities been capable of accepting an even larger diameter stage, a more efficient design could have been proposed. This thesis examines the most important components that need to be considered at the concept stage of an Australian launch vehicle project in order to avoid costly changes of direction entailing redesigns, expense and time.

3.2 BASIC VEHICLE CONFIGURATION

3.2.1 Launch Vehicle Mission As about half of all space launches are destined for geosynchronous (or geostationary) Earth orbit (GEO) and the Ariane commercial launch vehicle baseline mission is also to GEO, the baseline mission for an Australian Space Launch Vehicle (SLV) should also be chosen to be to GEO.

3.2.2 Launch Vehicle Concept While some may favour solid propellant, hybrid or pressure-fed liquid propellant vehicles, liquid propellant turbopump-fed vehicles are still by far the most common way to access space1 and so we take this as a system constraint to be accepted from the outset. Why Liquid Propellants?: Solid propellants have lower performances than do liquids. They need special, probably single-purpose, manufacturing facilities such as large mixers for single-pour casting and ovens for curing. The finished solid propellant boosters must be transported at least as a field assembly unit. For launch vehicles, this can be a very large and heavy exercise and probably means transporting explosives through a number of different states. As each state, federal and military authority has different explosives’ safety and transport regulations, this is a difficulty to be avoided. The alternative is to site the solid-propellant casting facilities at the launch site, as is done for the Ariane solid boosters which are cast at the launch site at CSG, Kourou. Hybrid Motors: Hybrid motors are preferable to solid propellant motors as only their fuels need to be transported in the rocket casing. Although these fuels are only inflammable and not explosive, they are still large and heavy. They have worse performances than do liquids but are marginally better than solid propellants. Their technology is still unproven but, in theory, they can be restarted and throttled in flight. Again, the transport problem can be solved by casting the fuel at the launch site. Pump-fed Engines are preferred to pressure-fed engines as the latter require large, heavy- walled tanks compared to those used for turbopump-fed liquid propellant engines. In large vehicles, the mass efficiency becomes low. Once again, their sizes and weights make them difficult to transport.

1 The new US Ares launch vehicle utilises existing solid propellant boosters because of the desire to escape the development of a totally new launch vehicle system. The second stage does however use a liquid propellant engine developed from existing hardware.

28

Chapter 3 System Considerations

Strap-on Boosters multiply the number of units that must be manufactured, transported and integrated into a vehicle. All must operate correctly together, in parallel. This makes a single-core vehicle preferable to a multi-core one. Expendable Vehicles: The reusable STS (space shuttle) was finally defeated by the fact that its operational cost was no lower than that of expendable vehicles. The high reliability required means that most of the vehicle has to be refurbished or, at least, inspected for each mission. This is particularly true for the thermal insulation tiles and the main engine turbopumps which have to operate at such high performance levels that they require maintenance after each flight. Large numbers of spare parts must be held in store in case they are needed. The external tank is expended and the solid boosters must be recovered and, at best, re-manufactured after each flight. In contrast, an expendable launch vehicle is newly manufactured for each flight and the parts and labour costs are accurately known and budgeted for in advance. Thus, their components need only be designed to operate in flight once.

Figure 3.1: Conversion of Ariane 44L to an optimised three-stage launch vehicle (HLV). This figure utilises material from an Arianespace source. (approximately to scale). Winged Re-entry: Wings impose a mass penalty that must be carried into orbit as well as large aerodynamic stresses on a launch vehicle. If the STS orbiter is out of alignment with the external tank and the flight path, by as little as 0.5 degrees, the aerodynamic forces at launch would break the orbiter from its mountings. On the other hand, the advantages of wings are the possibility of a wheeled landing on a standard runway and a larger cross-range on re-entry, thereby allowing a shorter wait time in orbit to access a given landing site.

29

Chapter 3 System Considerations

Baseline Launch Vehicle: In light of the architecture points discussed above, as well as the fact that it is arguably the world’s most successful commercial launch vehicle, the Ariane 44L (AR44L) vehicle is taken as the archetypical launch vehicle. It has been converted here into an optimised, three-stage vehicle (HLV=Hypothetical Launch Vehicle) without strap-on boosters but with the same payload capability as the AR44L. Figure 3.1 compares the two vehicles. For the conversion, the propellants, structure factors and specific impulses were assumed to be the same as those of the AR44L as were the payload, its fairing and adapter, and the vehicle equipment bay. These masses are shown in Table 3.2 [Ref 3.1].

Table 3.2: Payload and equipment masses used for both AR44L and HLV models Item Mass, kg Fairing 900 Payload 4568 Payload Adapter 200 Vehicle Equipment Bay 530

The “ideal velocity” (∆v) of the AR44L stack was calculated from published data [Refs 3.1, 3.2] and the ∆v of the HLV stack was constrained to give the same performance. The stage masses of the HLV were optimised to minimise the vehicle’s gross lift-off weight (GLOW). As a consequence of effectively reducing the size of the first stage by eliminating the four liquid strap-on boosters, the HLV vehicle requires that its second and third stages be considerably larger than those of the AR44L (221% and 316% respectively) while its first stage is about 53% of the size of the AR44L’s first stage and strap-on boosters combined. The resulting vehicle lift-off mass is some 124 tonnes lighter than is that of the AR44L. To avoid the need to develop two different engines for the first and second stages, the strategy, as used in the Ariane vehicles, is to use multiple engines in the first stage and a single engine in the second stage, both of the same design. As the second stage operates outside the atmosphere, the nozzle expansion ratio can be made larger than that of the first- stage engines which must be optimised for the average first-stage flight conditions. The sea level thrust to weight (T/W) ratio at lift-off in the first stage must be sufficiently high to minimise any wind effects [Ref 3.4]; for this reason, a minimum T/W ratio of 1.3:1 has been chosen. Similarly, the maximum axial load at burnout must not be excessively high in order to minimise loads and structural requirements on the user payload.

Table 3.3: HLV engine sizing data HLV Stage 1 2 3

Propellant UH25/NTO LH2/LOX No. of engines 5 1 1 S/L thrust, Mgf 92.6 ––––– ––––– Vacuum thrust, Mgf 103.8 103.8 24.4

Isp SL, sec 248.5 ––––– ––––– Ivac, sec 278.4 293.5 445.1 SL T/W ratio 1.3 ––––– ––––– Ignition axial load, g 1.3 0.5 Burnout axial load, g 3.4 1.8 2.6 Burn time, sec 110 222 720

30

Chapter 3 System Considerations

Analysis of the stage sizes leads to the choice of five engines for the first stage and a single, identical engine, with a nozzle extension skirt, for the second. The third stage has a cryogenic LOX/LH2 (Liquid Oxygen/Liquid Hydrogen) engine so as to obtain high performance leading to a lower GLOW rather than using the same propellant as in the first and second stages. Preliminary engine data are given in Table 3.3 above. As a baseline vehicle, the HLV is the first of a number of alternative vehicles that are compared with each other in this thesis. While being a single-core vehicle with no strap-on boosters, its similarity to its Ariane 44L parent vehicle is utilised. The total velocity budget (∆v) for the HLV launch is also assumed to be identical to that for the AR44L. The methods of calculation, and the vehicles of alternative and improved design, are discussed in subsequent chapters.

3.3 LAUNCH SITE LIMITATIONS

3.3.1 Earth’s Rotation If it is desired that a launch be undertaken from Australia, the economics of transport dictate that it should be from the mainland rather than from Tasmania so as to avoid the cost and time of moving components from the manufacturing sites on the mainland across Bass Strait to a Tasmanian launch site. This constrains the launch latitude to be between about 11°S in the extreme north of the continent to 39°S in the extreme south. This compares to 5°14’ North for the Ariane launch site at Kourou. Relative to the fixed stars (an inertial frame); the Earth rotates on its axis in approximately 23hrs 56mins. With a radius of 6378.137 km, a point on the Equator has a rotational speed of 465.12 m/sec. For a launch site on the Equator and a launch azimuth of due east, this velocity contributes to the velocity budget of the launch vehicle. Assuming a spherical Earth, the rotational velocity of launch sites at latitudes other than the Equator is given by:

Vrot = 465.12 Cos(φ) m/sec where φ is the latitude of the launch site. The further south an Australian launch site is located, the less the assistance provided by the Earth’s rotation to the overall velocity budget. Table 3.4 sets out the Vrot and the velocity loss at various launch sites compared to a launch from CSG, Kourou. Table 3.4: Earth rotational velocity at varying latitudes Launch site Earth’s Loss cf. CSG, Approximate Latitude Rotational Kourou, Location (Degrees) Velocity (m/sec) (m/sec) 5° N CSG, Kourou 463.3 0.0 11° S Cape York, Qld 456.6 6.8 17° S Cairns, Qld 444.8 18.5 23° S Rockhampton, Qld 428.1 35.2 28.5°S Gold Coast, Qld# 408.8 54.6 31° S Woomera, SA 398.7 64.7 33° S Newcastle, NSW 390.1 73.3 37° S Eden, NSW 371.5 91.9 39° S Wilsons Prom., Vic. 361.5 101.9 # Same latitude as KSC (Kennedy Space Centre, USA)

31

Chapter 3 System Considerations

Therefore, it is preferable, from the point of view of gaining as much Earth rotation assistance as possible, to have a launch site as close as possible to the Equator. However, for any location on the Australian mainland, the velocity loss is small, being only around 1% of the total velocity budget required.

3.3.2 Trajectory Inclination Much more important than the Earth’s rotational velocity is the minimum inclination of its orbit available for direct injection without a plane change. This is equal to the latitude of the launch site. For a GEO mission, the launch vehicle places the payload into an inclined transfer orbit (GTO) with its apogee on the Equator very close to geosynchronous height; this requires the perigee of the transfer orbit to also be on the Equator. The payload then separates from the launch vehicle and, at an apogee passage soon after release, fires its onboard propulsion system (or Apogee Boost Motor (ABM)). This increases the orbital velocity for circularising the orbit by raising the perigee to also be at geosynchronous height, as well as changing the plane of the orbit to be equatorial. The geometry of this situation is shown in Figure 3.2 below.

Figure 3.2: ABM firing velocity geometry

The standard Ariane 4 transfer orbit with perigee height of 200km has a perigee velocity of 10239 m/sec and an apogee velocity of 1598 m/sec, while the circular GEO, at a radius of 42165 km, has a velocity of 3075m/sec. Hence, the required velocity change at apogee is given by:

22 Sa −+=∆ 2 Sa Cosvvvvv φ

Commercially, we are interested in the payload that can be placed into GEO not that placed into the GTO/transfer orbit. Therefore, the calculation for the mass of the ABM that has to be added to the commercial payload uses the well-known rocket equation employed to derive the Ariane and HLV ideal velocities, viz:

∆ = vac mmIgv boi )/ln(.. with

mi = ms + mc +mp

mbo = mp + mc

32

Chapter 3 System Considerations where ∆ν = Delta-V, the velocity change required, φ = inclination of transfer orbit, g = acceleration due to gravity,

Ivac = vacuum-specific impulse of ABM propellant, mi = mass of payload and ABM ensemble before firing,

mbo = mass of ensemble at ABM burnout,

ms = mass of commercial payload, mc = mass of ABM case and

mp = mass of ABM propellant.

Assuming that the ABM used is a solid propellant motor with the same properties as the French Mage-2 motor, we have Ivac = 295.3 sec and (mc+mp)/mc = 13.25. The four equations above can be solved for the payload that the launch vehicle is required to deliver to the GTO/transfer orbit in order for the payload placed in GEO to be the same for various inclinations of the GTO/transfer orbit, i.e., the latitude of the launch site. The results are shown in Table 3.5. In all cases, the mass of the commercial payload delivered to GEO is 2681kg.

Table 3.5: Payload required to GTO for 2681kg delivered to GEO§ Apogee Mass to GTO Launch site Approximate Delta-V (kg) Latitude Location Required (Degrees) Total % (m/sec) 7° N CSG, Kourou* 1501.8 4768 100.0 11° S Cape York, Qld 1537.1 4835 101.4 17° S Cairns, Qld 1616.0 4989 104.6 23° S Rockhampton, Qld 1721.4 5203 109.1 28.5° S Gold Coast, Qld# 1836.5 5449 114.3 31° S Woomera, SA 1893.5 5578 117.0 33° S Newcastle, NSW 1940.9 5683 119.2 37° S Eden, NSW 2039.7 5916 124.1 § Includes payload adapter * Assumed Kourou latitude due to Ariane transfer orbit inclination # Same latitude as Kennedy Space Centre (KSC)

The rightmost column represents the percentage weight of a launch vehicle launched from Australia, equivalent to one launched from Kourou, to place the same payload in GEO. It can be seen that, at the highest southerly latitudes in Australia available to be chosen as a launch site, the payload capacity of a launch vehicle must be 24% greater than it would be for a similar vehicle launched from Kourou. To gain some idea of the implications of this for the worst case, compared with a launch site at Kourou, the 24% increase in mass equates to an approximately 8% increase in the linear dimensions for a launch vehicle and, compared with a launch from the Kennedy Space Centre (KSC) in the USA, it is about a 4.6% linear size increase.

33

Chapter 3 System Considerations

3.4 RANGE SAFETY

3.4.1 Range Safety Constraints and Considerations As previously discussed, the baseline mission for the HLV is to GTO (Geostationary Transfer Orbit). There are, however, a number of other missions that are likely to be of interest. Listed in a probable order of interest, they are: 1. Geosynchronous orbit (GEO); 2. Low Earth orbits (LEO); 3. Polar and sun-synchronous orbits; 4. orbits (ISS); and 5. Escape missions for solar system exploration. LEOs are of interest for scientific and military purposes. The most likely missions involve low inclination orbits for communications and Earth observation. Scientific LEO missions are likely to involve observations of the Australian continent and territories, while military missions are probably more interested in observations and communications of the areas relatively close to the Equator. Hence, for these missions, we can limit the orbit inclinations to, for example, 15 degrees for military missions and 43 degrees for scientific missions; the latter would provide observation coverage south to Tasmania. Polar and sun-synchronous orbits are of interest for both scientific and applications purposes and they are inclined within a few degrees of 90 degrees. The ISS orbit is inclined at approximately 51.6 degrees while Earth escape trajectories can be inclined variously depending on the particular mission. GEO, LEO and ISS directions are west to east; polar orbits are north to south or south to north while escape orbits can be in any direction, again depending on the mission. There are basically three problems associated with the launch track. Firstly, a clear area, in the order of ten kilometres, is required around the launch pad in case of a launch failure and any resultant fire and/or explosion. Secondly, the area under the ground track needs to be clear in case of an engine failure causing the vehicle to descend before reaching orbit. The instantaneous or walking impact point (WIP) will travel at least halfway around the planet before orbit insertion can be achieved and the area under this path must be clear of population and ground assets. Thirdly, the drop zones for the boost stages (aka stage fallout) must also be clear for a given radius around the predicted, most likely, landing point. Preliminary calculations indicate that the expended HLV stages will fall at approximately 500km and 2900km down-range from the launch site. When considering a location for a launch site in Australia, there are a number of choices that avoid most of the populated areas, viz: 1. Existing facilities at Woomera; 2. West coast for a continental launch path to the east or an oceanic launch path to the south; 3. South coast for a continental launch path to the north or an oceanic launch path to the south; 4. North coast for an oceanic launch path to the north or a continental launch path to the north; and 5. East coast for an oceanic launch path to the east.

34

Chapter 3 System Considerations

Tracks from launch sites other than the five listed above (such as a north coast site for an easterly launch), would cross inhabited areas of the Australian continent. Even if these areas could be cleared or the population warned, clear launch sectors are limited because of the presence of New Guinea and other Pacific islands. Similarly, launches from the north coast (choice 4) would track across the Indonesian archipelago if launched to the north. Although a less severe problem is encountered if the launch track is to the south across the Australian deserts, where the population density is very low, it would still present a problem to be dealt with. The same applies to choice 3 for a launch to the north. However, for both these continental launches, the first and second-stage drop zones would be over desert and the third-stage ignition would occur before the vehicle passed over the north coast which would minimise the probabilities of a launch failure occurring over foreign territories. Similar problems would be encountered for operational launches in any direction from Woomera. Until a few years ago, this site would, however, be suitable for developmental flights in which single stages or low numbers of complete vehicles were to be fired following the old Blue Streak tracks towards the north-west. However recently a large number of assets have been put in place close to the firing range, for example the “Prominent Hill” gold and copper mining venture lies just on the boundary of the Woomera range. Low frequencies of developmental test flights could be carried out but range safety would need to be assured by statistical methods taking into account “Failure Modes, Effects and Criticality Analysis” of the vehicles (FMECA). DSTO has recently been developing a state of the art range safety template based on these methods [Ref 3.4] and such templates would need to be applied to all flights to a suitable safety level. The most obvious and suitable choice for operational polar and sun-synchronous orbits is a south coast launch, tracking south across the open ocean towards Antarctica. However, the disadvantages of establishing a launch site for this purpose would be the difficulty of locating the tracking and telemetry stations and the difficulty of recovering the spent stages from the southern ocean, if required. When considering a south coast launch site, it can also be seen that, if the location was far enough west, ISS orbits would be within reach of a launch track to the south-east completely over water, thereby avoiding Tasmania and New Zealand. For GEO and LEO launches, one would like the launch site to be as far north as possible in order to access as many orbit inclinations as possible without a plane change being required. A launch from the west coast, eastward across the Australian deserts, is one possibility as is a launch from the east coast across the Pacific Ocean. The east coast solution is, of course, the one that is most often considered. However, the Pacific islands of Melanesia present the problem of population centres obstructing a clear launch trajectory and the ecosystem of the Great Barrier Reef is also an issue. The most prominent, previous launch site proposals have concentrated on the north end of the eastern seaboard. A launch sector originating from the coast south of, for example, Gladstone in Queensland, was suggested because it is free of most island obstacles until near Eden in New South Wales when the proximity of New Zealand becomes a problem for much of the launch sector. It is, at this stage, after considering the map of Australia and the Pacific Ocean, that one comes to the conclusion that a space launch program in Australia would have to bite the bullet and accept that there would be a requirement for two launch sites, one on the east coast for an easterly launch sector and one on the south coast for a southerly launch sector.

35

Chapter 3 System Considerations

3.4.2 Drop Zones and Azimuth As well as considering the launch track for the nominal (to GEO) mission, there are missions of differing orbit inclinations that can be accessed by changing the launch azimuth. As mentioned earlier, the minimum inclination available is achieved by a launch with an easterly azimuth. Lower inclination orbits must be accessed by a plane change when nearing the Equator. Orbits with a greater inclination can be accessed by increasing or decreasing the launch azimuth for the easterly launch. It is, therefore, important to examine all launch azimuths when selecting a launch site. The east coast of Australia suffers from the down-range existence of New Caledonia, New Zealand, numerous Pacific atolls and other minor land masses, many of which are inhabited. However, it is fortunate that, at the drop zone of the first stage (~500km) and the (~800km), the east coast is bordered by open ocean. So, the land and populations in the impact range of the second stage (~2900km) are of most concern which is one of the considerations that will limit the launch azimuths. The WIP, at any time throughout the launch, will travel at least halfway around the globe before orbit is reached. The ground track after the second-stage drop zone is, therefore, also of concern. Third-stage ignition will have occurred in this region and will have been burning for some time, as well as the flight being outside the atmosphere, so that aerodynamic forces and heating would then be zero. This means that there would be fewer contingency factors to indicate a possible launch failure and impact before orbit is reached. The drop zone ranges quoted above are for the baseline HLV. The stage drop zones will be modified if the design of the vehicle is changed. In addition, they can be pushed or pulled to shorter or longer ranges by exchanging masses between stages. This will, however, be at the expense of a lift-off weight penalty depending on the movement of the drop zone sought and the size of the exchange required. A subsequent chapter in this thesis examines alternative vehicles and the position of the drop zones for the vehicle stages and payload fairing. The launch sectors for the recommended southern launch site for high inclination launches is over ocean completely clear of land masses until the Antarctic continent, which is virtually uninhabited, is reached.

3.4.3 Australian Space Licensing In December 1965, the Australian Department of Supply released a study report which proposed a launch site be established near Darwin for an advanced concept ELDO launch vehicle [Ref 1.8]. ELDO decided however, to adopt the French equatorial launch site at Kourou for its future operations. European launch operations in Australia ceased and ELDO and ESRO ultimately became ESA. As a result of a suggestion by this author, a study was commenced in 1986 into the feasibility of establishing an international spaceport on Cape York Peninsula in Queensland. The concept was to provide, on a commercial basis, the ground facilities for any international organisations who wished to utilise the near-equatorial location for their own launch vehicles. The feasibility study and subsequent scoping studies [Refs 1.9-1.14] were funded by the Queensland Government and carried out by the Institution of Engineers’ Australian National Committee on Space Engineering. The outcome was that a spaceport was not established but interest was reawakened in its possibilities. Several other spaceport studies followed but, more importantly, as a result of

36

Chapter 3 System Considerations a Senate inquiry and the recommendations of Australian and overseas advocates of launch services in Australia, the Australian Parliament recognised the need for a legislative framework to regulate civilian space activities in Australia, including by Australians outside Australia. As a result, the Space Activities Act 1998 (“the Act”) [Ref 3.5] came into being and the Space Activities Regulations 2001 (‘The Regulations”) [Ref 3.6] under this Act were developed. Australia is a signatory to the five UN space treaties which include provisions that place direct responsibility (unlimited liability) on the Commonwealth for damage caused to other countries and their nationals by rockets (privately or publicly owned) launched from Australia. The three treaties of most significance for Australia, as a space launching nation, are: • The Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and other Celestial Bodies; • The Convention on International Liability for Damage Caused by Space Objects; and • The Convention on Registration of Objects Launched into Outer Space.

The Liability Convention imposes on signatory nations international liability for damage caused to persons or property in third-party nations by all ‘space objects’ launched from the nation-state’s territory/facility, or whose launch was procured by the nation, its agencies or non-governmental entities. In addition to the liability provisions, nations are also responsible, under Article IX of the UN , for harmful contamination of, and any adverse changes to, the Earth’s environment. In the case of a launch from Australia, the Commonwealth would be held responsible whether or not it was itself involved with the launch. The Act and Regulations require that every flight that travels more than 100km above the Earth’s surface should be licensed. The licensing is administered by the Space Licensing and Safety Office (SLASO) of the Australian Department of Innovation, Industry, Science and Research. SLASO has been established to take responsibility for implementing the regulatory and safety regime for space activities in Australia and by Australians outside Australia. It has responsibility for enforcing the provisions of the Act and the Regulations as well as any Agreements entered into by the Commonwealth for space launch projects. SLASO is headed by a Director who, advised by expert assessors, is responsible (under delegation from the Minister) for the licensing approvals of space launch facilities, space launches and associated activities. This implies that Australian nationals who own, or who have a controlling interest in, foreign companies which produce space technology components, sub-systems or complete spacecraft, or who own the components destined to be orbited above 100km, must be licensed2. SLASO assesses applications for approval under the Act and has responsibility for the ongoing development of subsidiary regulations and guidance material. This ensures that space activities do not jeopardise public safety, property, the environment, Australia’s national security, foreign policy or international obligations. The Act also ensures that

2 It should be questioned as to whether Australian nationals working for a foreign space technology company whose product is launched into space must be licensed as well?

37

Chapter 3 System Considerations regulated activities are adequately insured for third-party purposes, and that accidents are investigated. Under the Act, certain space activities are prohibited unless appropriate approvals are obtained, for example, a licence is required to operate a space launch facility. Special, relaxed rules can be applied to Australian approved scientific and educational organisations [Ref 3.7].

3.4.4 Australian Flight Safety Code The Space Activities Act, Regulations and documents setting out the rules pertaining to flight safety [Ref 3.8] are available on-line from the SLASO web site. There are also documents describing the method used to calculate the “maximum probable loss methodology” [Ref 3.9] to be applied when implementing the safety guidelines, and listing designated protected assets [Refs 3.10, 3.11] to be considered in the application of the safety code. At the time of writing, SLASO recognises two applications from launch proponents for the consideration of permits and Figure 3.3, adapted from the SLASO web site, shows the indicative launch corridors of and the regional assets of potential interest to, those proponents. The map also shows designated and protected oil facilities that are afforded special protection under the space safety regime [Ref 3.12]. As the Flight Safety Code and Maximum Probable Loss Methodology would need to be applied to any launch vehicle designed in Australia, a launch site’s location, mission, vehicle design and trajectory would have strong interactions with each other in the safety regime. Subsequent chapters of this thesis will discuss possible launch corridors and their safety implications for the HLV launch vehicle.

38

Chapter 3 System Considerations

Figure 3.3: Australian space activities’ regional map showing indicative launch corridors and associated protected assets for two recognised launch proponents. Map is reproduced and modified from SLASO web site [Ref 3.12].

39

Chapter 3 System Considerations

3.5 GUIDANCE AND TRACKING Modern launch vehicles rely on self-contained inertial guidance systems (INS) or global positioning systems (GPS) to follow pre-programmed launch trajectories. However, continuous radio communications must still be possible between the ground and the launch vehicle. This is required for several reasons: 1) Radar tracking for trajectory determination; 2) Range safety requirements for possible flight termination; 3) Telemetry availability for possible diagnostic use; 4) Command override for critical operations (e.g., spacecraft deployment); and 5) Continuous telemetry and voice communications for manned flights. In order to satisfy the requirements for Telemetry, Tracking and Control (TT&C), a number of tracking stations must be established along the launch corridor. Firstly, a tracking station must be located at each launch site. Down-range stations must also be established to determine the trajectory of each stage and, thereby, its fallout point. In addition, the trajectories of the third stage and the spacecraft need to be determined, as well as the receipt of telemetry, in order to ascertain the status of the release sequences of the payload spacecraft. A south coast launch site for polar and sun-synchronous launches has the advantage of no land masses being to the south until Antarctica is reached so range safety is a minimum consideration. But the disadvantage is the need for a ship-borne tracking system as the second-stage fallout occurs well before the Antarctic coast. A tracking station located at one of the Australian Antarctic bases, Casey or Davis, and relayed by a permanent satellite communications channel would serve to give TT&C information for these orbits. However, the effect of the time delay due to satellite communications would need to be considered for critical operations. For launches from the south coast to ISS orbits, a ship-borne tracking system would probably be needed as the launch corridor is well south of the Australian continent until the second-stage fallout south-east of Tasmania. The south coast of Tasmania and/or islands to the south of New Zealand could be utilised for the location of a tracking station. For equatorial and low inclination orbits launched from an east coast site, there are many islands on which a tracking station could be based. Some are uninhabited (and actively volcanic) but could be used as a base if ship-borne tracking was still required. The exact location of tracking stations would need to be determined by a more detailed study than this thesis and would depend on the trajectory required. The provision of a ship-borne system would be expensive and should be avoided if possible. However, a transportable system that could be moved depending on needs is one solution to the TT&C problem. Launch corridors and TT&C locations are examined in more detail in subsequent chapters of this thesis.

3.6 OPERATIONS A launch vehicle system requires a considerable amount of initial design effort along with a suite of facilities for development and testing. These facilities are needed not only for vehicle development but also for ongoing design improvements and incremental design

40

Chapter 3 System Considerations upgrades. There are, therefore, up-front costs for the construction of these facilities and for vehicle development. These are often seen by financiers as the major costs of a space launch system and are stumbling blocks when requiring support for a program from governments [Ref 3.13]. However, about 60% of all life-cycle costs [Ref 3.14] are for the operational and support expenses associated with the launch campaigns (Figure 3.4). The longer the operational life of a launch vehicle, the greater the proportion of the program costs it requires.

Figure 3.4: Costs of program life-cycle phases. Figure adapted from SMAD [Ref 3.14]

As these costs can be charged out to the launch customer, in the form of launch charges, they are of great importance when considering operations and support at the time of concept development. These are areas in which major program cost savings can be achieved. For a commercial launch system, this translates to competitiveness, market share and profit. As historically, 75% of system constraints are locked in during the system concept phase [Ref 3.15], it is of vital importance to consider the operational and support phases as prime requirements in the concept definition.

3.6.1 Transport The transport of components and personnel can consume a great deal of the program’s costs and can place constraints on the system’s design. The Ariane launch vehicles are a case in point. The largest component, the airframe for the Ariane first and the third stage, was manufactured by Aerospatiale in Toulouse, France, while the second stage was manufactured by MBB in Bremen. The individual components were transported by road, rail and sea to a major European harbour where the complete launch vehicle was placed in an air-conditioned “coffin” and loaded aboard a ship. Accompanied by an attendant to assure benign environmental treatment during its trip, it was transported by sea to CSG at Kourou, French Guiana. The political implications of the principle of “Juste Retour”, whereby participating nations are ensured a fair return of work for their financial contributions, required that components were moved around Europe as part of the manufacturing process. The Ariane program suffers from the further cost disadvantage of the location of the CSG launch site. As it is at

41

Chapter 3 System Considerations the end of a sea from Europe, technicians have to be transported and accommodated at a site remote from their home base and, presumably, provided with a per diem allowance, thereby further increasing operating costs. A launch site on mainland Australia, located near the major industrial centres of Melbourne, Sydney, Brisbane or the administrative and government centre of Canberra, would not suffer the same transport difficulties as those of the European program. Situating a facility for manufacturing the large stage propellant tanks at the launch site would alleviate any of the difficulties associated with the transport of oversize components.

3.6.2 Launch Campaign One of the major measures of system performance and commercial viability is “Operational Availability” or “Responsiveness” [Ref 3.16]. For a launch vehicle, this means reliability during the launch phase but, also, freedom from technical problems during stacking, integration and countdown on the launch pad. Flexibility to customer orbit requirements and launch environment must also be considered. The launch needs to be able to reliably lift off on time within a specified launch window. A simple and short sequence of stacking, integration and check-out minimises the time that technicians need to be employed at the launch site while maximising pad availability for subsequent launches by decreasing turnaround times. Again, it is seen that operations, support and logistics are of prime importance in ensuring a commercially successful launch system.

3.7 LOADS AND AERODYNAMIC HEATING

3.7.1 Maximum Dynamic Pressure

The maximum dynamic pressure (Qmax) occurs over a range of altitudes depending on the trajectory of the vehicle. For most vehicles, the altitude range is between approximately 9000m and 11000m altitude with Qmax occurring at the upper end of the range for typical space shots. As the greatest wind shears are between 9000m and 12000m, the maximum bending loads are also seen in the same region.

3.7.2 Heating The maximum heating rate and the total heat input during flight through the atmosphere are also dependent on the trajectory. Their values are necessary in order to calculate the properties required for the structural materials plus the amount of insulation needed to protect the structural elements and electronics and to insulate the pressurant gases and cryogenic propellants against undue boil-off. The main heating concerns are, of course, during the first-stage flight. However, some heating also occurs during the second stage which may place constraints on the jettison fairing, for example, the nominal time for the Ariane 4 fairing of the jettison is specified to be at an aerothermal flux not exceeding 1135 W/m2 although this requirement can be modified by the launch user.

3.7.3 Structural Loads As well as the normal loads due to acceleration, hydrostatic loads, aerodynamics, wind shear and other forces, some missions impose much higher loads than does the nominal mission. In all orbital missions, the trajectory requires a pitch-over during atmospheric flight. However, for low Earth missions, the pitch program may need to be much more aggressive than it is for the mission to GEO.

42

Chapter 3 System Considerations

The Ariane flight profile to GTO is lofted above the transfer orbit perigee height and approaches the transfer orbit from above the perigee. A mission to circular LEO may, however, require that the orbit is approached from below. This implies a much more rapid launch pitch-over requiring a greater angle of attack, thereby inducing larger aerodynamic forces on the airframe. It is, therefore, necessary to identify all missions before a vehicle design is started in order to account for the worst possible scenario for the structural forces. Having designed the nominal trajectories and proceeded to a vehicle design, it may be found that it is then necessary to modify the trajectories in order to satisfy any heating and structural load constraints that cannot be handled in the design. Fortunately, as far as system design is concerned, these factors are second order effects and shouldn’t influence the basic system concept, especially the critical logistics elements.

43 Chapter 4 Strategic Propellant Selection

4.1 IMPORTANCE OF CORRECT PROPELLANT SELECTION Launch vehicle propellant selection is one of the major design constraints that must be chosen correctly at pre-concept stage. It is normally associated with vehicle performance but its choice should be made on more strategic grounds. As an ongoing consumable any operational impacts of the propellant choice will continue throughout the life-cycle of the launch vehicle project. As a material; manufacture, handling and storage facilities must be designed, manufactured and installed at a number of sites, viz: launch site and engine development and test sites. Any change in propellant selection will require changes, possibly major, with associated costs, in these facilities. Launch vehicle and engine design is intimately affected by propellant selection and a change may mean a total redesign and redevelopment of much of the launch vehicle and associated launch site facilities. It may also mean the development of a totally new technology. Propellant choice can also have project strategic affects that may mean the difference between project success and failure. Correct selection is therefore an important case of risk management within a project Many obstacles can be placed in the path of a project such as the development and operation of a space launch system. Examples include: politics (federal, state and local), bureaucracy, import/export licenses/restrictions, missile technology control regime (MTCR), Hague Code of Conduct against Ballistic Missile Proliferation, US International Trade in Arms Regulations (ITAR), National Security, Environmental issues and OH&S. It is therefore of strategic importance to the project to make decisions that avoid as many of these obstacles as possible. The choice of propellant is a major decision that could be criticised on many grounds. A launch vehicle system is no more than a few buildings at a launch site and a light weight aerospace vehicle mainly of stainless steel, aluminium and carbon fibre. It is when the highly energetic propellants are added to the mix that criticisms and obstructions can be made which will add to the risk of project failure or at very least increased costs and delays. Nearly two-thirds of the recurring cost of providing a space launch is expended in the activities that take place at the launch site to prepare for and support the launch [Ref 4.1]. It is therefore in the ongoing operations activities that the greatest savings in cost can be achieved that will help to ensure the success of a commercial launch vehicle project. While constraints are imposed by the launch sites, manufacturing facilities and transport available, the choice of vehicle propellant is one constraint that crosses many disciplines, continues throughout the life cycle of the vehicle system and affects the entire launch system. Luckily or unluckily depending on the point of view, there are only a small discrete number of choices that can be made for the propellant to be used in any proposed launch vehicle. As a continuing consumable needed for engine and stage development, testing and calibration firings as well as for flight, the cost of the propellant must be as low as possible. The safety factors involved in the manufacture, transport, storage and handling must be as high as possible as well as toxicity being low. Transport by road, rail, sea and air must be possible without excessive safety precautions being required. The feedstock

44

Chapter 4 Strategic Propellant Selection for manufacture of the propellant must be plentiful and easily obtainable. This implies that a domestic source would be preferable for political and cost reasons. If not available from a domestic source the propellants should be of common industrial use to avoid importation problems caused by such obstacles as the missile technology control regime. An established economy utilising the propellant is almost essential as this will mean that the technologies for production, storage, transport and handling will not need to be developed. Dual sourcing is also desirable for surety of supply. Manufacture should be simple and the characterisation of the propellant must be consistent. The physical properties of the propellants must be satisfactory, e.g. high density, low freezing point, high boiling point, low viscosity and high thermal conductivity.

4.2 HISTORICAL PROPELLANTS AND THEIR PROBLEMS From the beginning of spaceflight the choice of propellant was dominated by the needs of the cold war and the development of ballistic missile systems. Consequently, performance was the prime consideration. Recent trends have however been tempered by the need for “Greenness”. Low toxicity, safe and easy handling and low pollution is now a very desirable characteristic and internationally the hunt is on for environmentally benign propellants that still have high performance figures [Ref 4.2]. While every conceivable combination of propellant fuels and oxidisers have been considered and tried in developmental and production engines, there are basically only five combinations which have been commonly used in launch vehicles. These are listed in Table 4.1.

Table 4.1: Propellant Combinations and Example Vehicles Example Approx Oxidiser Fuel Vehicle(s) Ivac, sec Date [Ref 4.3] LOX Alcohol (Ethanol) V2 239 1944 Redstone 265 1955 LOX Kerosene/RP1 Atlas-D 282 1960 Blue Streak 282 1960 Saturn-V 304 1967 LOX LH2 Saturn-IVB 421 1967 Ariane 446 1979 N2O4 Hydrazine(s) Titan-2 296 1962 Ariane 296 1979 H2O2 Kerosene Black Knight 265 1958 Black Arrow

4.2.1 The Good 1 Liquid Oxygen, O2, LOX: Much experience with the most efficient of the oxidisers , liquid oxygen, has shown itself to be quite tractable in handling with minimal safety problems and is completely “Green” with no associated environmental problems. It is also relatively low cost and can be produced in Australia in the quantities needed for a launch vehicle program. This would be the oxidiser exclusively recommended for launch vehicle application.

Liquid Hydrogen, LH2: Hydrogen’s exceptional performance figures place it in the good category. Because of performance requirements, LOX/LH2 is almost mandatory as the

1 Propellants containing fluorine, chlorine and boron are not considered, being too expensive, polluting and toxic [Ref 4.21].

45

Chapter 4 Strategic Propellant Selection propellant for upper stages. Luckily there are no toxicity problems with this material. There are, however, several severe drawbacks to hydrogen as a fuel. The worst of these is its very low density, requiring large tanks and off optimal fuel lean mixtures. Liquid hydrogen also must be kept at extremely low (cryogenic) temperatures. It causes hydrogen embrittlement of some materials and the small size of the molecule means that it has a tendency to leak through materials, especially small cracks in welds. Leaks can ignite and burn with an invisible flame. This makes handling extremely difficult but all this is overridden by its high performance. It is available from steam reforming of petroleum and gas resources as well as from biological sources and electrolysis of water [Ref 4.4].

4.2.2 The Bad Kerosene/RP1: Following the use of Ethanol as a fuel, designers moved to kerosene which provided higher performance. Aviation fuels such as JP-4, JP-5 or Jet-A1, however, had problems in that they were not specified to the standards required by rocket engines. Raw kerosene used as coolant would suffer cracking and gum formation. Lightweight products in the form of gas bubbles, and heavy ones in the form of engine deposits, then blocked the narrow cooling passages. The coolant starvation raised temperatures further, accelerating breakdown. In 1953 Rocketdyne commenced REAP (Rocketdyne Engine Advancement Program), one outcome of which was the development of a special grade of kerosene (RP1) suitable for rocket engines that met specified requirements of density, heat of combustion, aromatic content and low sulphur (<20ppm). RP1 (MIL-R-25576) is a narrow cut fraction of petroleum which must be refined from crudes with high napthene content [Ref 4.5]. It is subjected to further processing to remove undesirable compounds such as sulphur containing hydrocarbons. Russian propellant chemists developed similar fuels, T(S)-1 and RG-1. In addition they developed synthetic (nondistillate) kerosene called “Sintin” [Ref 4.3]. By adding strain to the molecular bonds the heat of formation was increased along with the density. An increase of several seconds of specific impulse was achieved in this manner. It is believed that the Sintin production plant is now no longer in use, probably due to the high cost of production. The increasing scarcity of crude oil, in particular that with high napthene content, the cost of refining or synthesising, the need to characterise each batch and the fact that it is a single use product are amongst the drawbacks to the use of RP1. It has been, however, a very successful fuel in particular for lower stages (e.g. Saturn 5). The sulphur content of RP1 contributes to corrosion in engines so recently a REAP-2 program has been investigating the properties of RP2, an ultra-low sulphur version of RP1 for use in reusable launch vehicles where corrosion becomes more of a problem than for expendable engines. From the Australian standpoint a refinery would need to be constructed or modified to produce this fuel so the economics of production of this single use product in this country could not be justified. RP1 would thus need to be imported which is undesirable from a strategic point of view. Importation would be an ongoing need and the political situation could change midway through the project or worse, nearing the end of equipment development. The supply and manufacture problem as well as the need to characterise each batch and the relatively high carbon emissions from the exhaust (running fuel rich) are the reasons why this fuel has been placed in the “Bad” category.

Hydrogen Peroxide (HTP, High Test Peroxide), H2O2: Hydrogen peroxide has been successfully used as a launch vehicle oxidiser by the British in their Black Knight sounding rocket and its successor the Black Arrow launch vehicle. Its performance with kerosene

46

Chapter 4 Strategic Propellant Selection

(Isp=276s) [Ref 4.6] is about forty seconds of specific impulse less than liquid oxygen (Isp=301s) and this is its main drawback. It is also extremely corrosive in the high concentrations (75% +) required for launch vehicle oxidiser use. It is very difficult to handle as it decomposes into water (steam) and oxygen in the presence of impurities. Conversely this makes it useful as a for spacecraft. It is not manufactured in Australia and must be imported. See [Refs 4.7, 4.8] for further discussion.

4.2.3 The Ugly

Hydrazine, N2H4: Hydrazine is the worst of the fuels in that it is toxic and carcinogenic as well as being extremely expensive. Its replacement is one of the main reasons for the international search for “Greener” fuels. Hydrazine and its compounds are strong reducing agents, first used as a rocket propellant (called B-Stoff) in the Messerschmitt Me-163 Komet. They are most famous in the west for fuelling the Titan-II and Ariane 1-4 launch vehicles. It was also the fuel for the Russian SS-18 Satan ICBM as well as for many other engines. Hydrazine is hypergolic (spontaneously ignites) with many organic materials including nitrogen tetroxide as oxidiser. Hypergolicity is one of hydrazine’s advantages as is its relatively high density and its stability meaning that it can be held tanked on-board for relatively long periods waiting for launch windows. Hydrazine as a rocket propellant is normally used with one of its related compounds, Mono-Methyl-Hydrazine (MMH) and Unsymmetrical Di-Methyl Hydrazine (UDMH) (MIL-D- 25604). Titan-II (and Titan III & IV) used Aerozine-50, a 50/50 mix of hydrazine and UDMH which is more stable than hydrazine and has a higher density and boiling point than UDMH alone. Ariane 1-4 used UH25 a mix of UDMH and 25% hydrazine. Hydrazine is also used as a monopropellant in Attitude and Orbit Control Systems (AOCS) and for this purpose it probably has no replacement except for hydrogen peroxide, which has its own problems. In the USA the price in 1959 for drum quantities of hydrazine was under $7.00 per kg but due to its highly poisonous and carcinogenic nature, environmental regulations pushed the price up and by 1990, NASA was paying $17.00 per kg [Ref 4.9]. Hydrazine compounds are currently available in bulk from China [Ref 4.10] but in Australia the environmental movement has vowed to oppose any launch vehicle fuelled by hydrazine.

Nitrogen Tetroxide, N2O4: Nitrogen Tetroxide or NTO is every schoolboy’s favourite rocket oxidiser. Its manufacture is even on the NSW HSC school practical chemistry curriculum. Its use as a rocket oxidiser succeeded nitric acid and the red and white fuming nitric acids (RFNA, WFNA - also beloved of schoolboy chemists). It is the highest performing Earth storable liquid oxidiser when combined with hydrazine based fuels. It is highly toxic and extremely corrosive hydrolysing to nitric acid. Along with hydrazine it has been the cause of the highest proportions of launch site spills and accidents [Ref 4.11]. Relatively easy to manufacture and low in cost, it is also relatively stable at low temperatures and can be transported relatively easily but not by air or mixed with passengers by rail. It is also hypergolic with many fuels which simplifies ignition. NTO is considered “semi-green” as it is non-carcinogenic. In 1959 one-ton cylinders of NTO cost $0.15/kg but by 1990 NASA was paying $6.00/kg again due to environmental regulations [Ref 4.12]. Whether Australian environmental groups would oppose the use of NTO is unknown. This is probably the main factor determining whether NTO could be promoted from “The Ugly” group to “the Bad” or “The Good” group. Whether the manufacturing and handling difficulties of NTO are greater than those of H2O2 is probably only a small matter of degree. The decomposition products of H2O2 are at least completely non-toxic whereas the nitric acid that can result from NTO hydrolysis is not. In relatively small quantities, such as that required for AOCS subsystems on payload spacecraft, NTO could probably be classed as an acceptable oxidiser.

47

Chapter 4 Strategic Propellant Selection

4.3 FUTURE PROPELLANTS Increasing concerns over safety and environmental factors have prompted a search for propellants that are less toxic, easier to manufacture, transport and store and have cleaner exhaust emissions [Ref 4.2]. The principal desire is to eliminate the use of hydrazine based compounds because of their carcinogenic risk. This impacts both boosters and spacecraft and bipropellants.

4.3.1 Performance Considerations for Launch Vehicle Fuels Alcohols: The archetype of space launch vehicles, the German V2/A4, used Ethanol derived from the potato crop as a fuel, not for performance reasons but for strategic reasons. It was believed at the time that petroleum based fuels would be difficult to obtain due to other priorities so Ethanol was chosen instead [Ref 4.13]. The specific impulse achieved in the V2 was relatively low (Ivac=239s, SL Isp=203) due to dilution of the ethanol with water for chamber cooling purposes and a loss of some 10-15s of Isp from high frequency combustion instabilities that could not be detected with the instrumentation of the time. The descendent of the V2, the US Redstone, managed a higher Ivac of 265s (SL Isp=235s). Current technologies could achieve even higher performance. Figure 4.1 below shows relative theoretical specific impulse performances of the first five alcohols. Propellant Chemistry - A Structural Viewpoint: The addition of the hydroxyl radical to an alkane to form an alcohol (for example methane > ) partially oxidises the alkane so that the energy available for combustion is lower with the alcohol than with its corresponding alkane. The specific impulse figures for alcohol are thus lower than for the alkanes. As the chain length of the backbone increases in successive alcohols, however, the effect of the hydroxyl becomes relatively less and the maximum specific impulse increases. From figure 4.1 it can be seen that there is a considerable increase in energy (heat of formation) from methanol to ethanol and the trend continues with specific impulse increasing up to the plotted values for n-pentanol.

In the case of the corresponding Alkanes, starting with Methane, the addition of further -CH2 radicals extends the backbone to form ethane, propane, butane and higher alkanes respectively. The H:C ratio starts at 4:1 for Methane and drops to 3:1 for Ethane, 8:3 for Propane, 5:2 for Butane and then approaches 2:1 as the chain increases in length. The heat of formation per carbon atom therefore decreases with increasing Alkane chain length. The decreasing specific impulse2 of the LOx/Alkane combination approaches the increasing specific impulse of the Lox/Alcohol combination as the chain length increases. A structural view can assist the engine designer to visualise the effects of changing the propellant choice. Table 4.2: Properties and performance figures of alcohol fuels at engine (first stage) design conditions of figure 4.1 Freezing Boiling Max Specific Propellant Formula Point, C Point, C Isp Gravity Methanol -97.0 64.7 318 0.792 CH3OH

Ethanol -114.3 78.4 324 0.789 C2H5OH

n-Propanol -126.5 97.1 328 0.803 C3H7OH

n-Butanol -89.5 117.7 330 0.810 C4H9OH

n-Pentanol -77.6 138.0 331 0.814 C5H11OH

2 At the mixture ratio giving maximum Isp

48

Chapter 4 Strategic Propellant Selection

Methane (CH4) and the Lower Alkanes (paraffins): Methane combined with oxygen gives a maximum specific impulse in the 370s range. It has a greater density (s.g. 0.423) than hydrogen (s.g. 0.071) but lower than RP1 (s.g. 0.81). It has high thermal conductivity and does not soot or coke in cooling channels and injector elements [Ref 4.14]. It is the greenest of the hydrocarbon fuels having a Hydrogen:Carbon ratio of 4:1 while longer chain hydrocarbons approach an H:C ratio of 2:1. Ethane, propane, butane, pentane and some of the longer chain alkanes, alkynes and alkenes are also suitable as fuels with decreasing performance but increasing density with increasing chain length. However multi and cyclic bond compounds should generally be avoided as they have a tendency to polymerise at the high temperatures found in regenerative cooled turbopump driven engines.

Performance of LOX/Alcohol Propellants 340

330

320

310

300

290 Methanol 280 Ethanol Propanol 270 Butanol Theoretical Vacuum Specific Impulse, sec Impulse, Vacuum Specific Theoretical Pentanol 260 012345 Mixture Weight Ratio, O/F

Figure 4.1: Theoretical vacuum specific impulse for LOX/n-Alcohol propellant combinations. Chamber pressure is 10MPa, Expansion ratio, ε=14. Other engine parameters are typical first stage values. Curves computed using NASA Glenn CEA program [Ref 4.15].

While all the lower alkanes can be considered as potential fuels, propane is of particular interest. Its vapour pressure near the boiling point (-42ºC) is several atmospheres which would require heavy high strength tanks to store at those temperatures, however its freezing point is lower than that of liquid oxygen and it can thus be stored at chilled temperatures near the LOX boiling point allowing the use of a common bulkhead between the tanks with the consequent savings in space and weight. At these temperatures the density of liquid propane is considerably increased to (SG = 0.790) [Ref 4.8] and the vapour pressure is reduced to about 0.1mm Hg. The loss of performance from methane to propane is approximately 10s of specific impulse but its density is nearly double that of methane. Hydrogen: The performance figures for Oxygen/Hydrogen are significantly higher than any other propellant combination1. As can be seen from the curves in Figure 4.2 the theoretical specific impulse performance can approach 500s. As has been mentioned before the drawback with this propellant is the low density of liquid hydrogen. However for performance reasons

49

Chapter 4 Strategic Propellant Selection

Hydrogen and Alkanes as Fuel - Performance with LOX as Oxidizer 550

500

450

400

ρ ρ + r)1( Methane 350 ρ = fo av r + ρρ Ethane of Propane 300 Butane Pentane 250 Hydrogen-14

Theoretical Vacuum Specific Impulse, sec Impulse, Vacuum Specific Theoretical Hydrogen-100

200 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Oxidizer/Fuel Mixture Ratio by Weight

Figure 4.2: Theoretical Vacuum specific impulse for LOX oxidiser comparing various alkanes and hydrogen at two expansion ratios (ε = 14 and 100) as fuel. Chamber pressure is 10MPa. Curves computed using NASA-Glenn CEA program [Ref 4.15] many launch vehicles adopt LOX/LH2 as the propellant for the third stage. For example the Ariane-44L third stage (H10-3) used this propellant combination in the HM7B engine achieving a vacuum specific impulse of 445.1s with a gas generator cycle running at a chamber pressure of 3.5MPa, a mixture ratio of 4.77:1 and a nozzle expansion ratio of 62.5:1 [Ref 4.16]. The reason for this third stage propellant choice will become obvious when the impact of alternative propellant choices on vehicle size for a given performance is examined in chapter 6

Effect of Density on Propellant Choice: The bulk (average) density of a propellant (oxidiser plus fuel) as a function of mixture ratio is given by [Ref 4.17]: ρ ρ + r)1( ρ = fo av r + ρρ of and the density specific impulse is given by:

= ρ II savd where, ρav = the average density of the propellant ρo = the average density of the oxidiser ρf = the average density of the fuel, and r = the mixture ratio, oxidiser mass/fuel mass Id = the propellant bulk specific impulse Is = the specific impulse Specific impulses for LOX with methane, propane and RP1 are reproduced below in figure 4.3 while the propellant properties are given below in table 4.3.

50

Chapter 4 Strategic Propellant Selection

From the figure it can be seen that on a pure performance (specific impulse) consideration Methane is the best performer followed by Propane and then RP1. However when another performance indicator, the propellant “density specific impulse”, is examined the situation is changed. Figure 4.4 plots the density impulse for the three fuels under consideration while Table 4.4 set out the values. The data in figure 4.4 and table 4.4 indicate that both Propane and RP1 have their maximum density impulse and maximum specific impulse at nearly the same mixture ratio. Methane on the other hand has widely differing mixture ratios for the two maxima. Choosing a methane mixture ratio to maximise the specific impulse will lead to a low propellant bulk density penalising the structure in the form of larger hence heavier tankage. Conversely, choosing to maximise the density impulse will lead to a fuel lean mixture that will compromise the specific impulse (exhaust velocity) with only a small increase in bulk density. Table 4.3: Propellant Properties Freezing Boiling Specific Propellant Comment Point, K Point, K Gravity Liquid Oxygen 54.75 90.20 1.140 At boiling point Liquid Hydrogen 14.01 20.28 0.071 At boiling point Liquid Methane 90.67 109.15 0.423 At boiling point 0.583 At boiling point Liquid Propane 83.15 231.15 0.790 At LOX boiling point RP1 mixture mixture 0.810 At 20C As there are structural weight overheads involved in any stage design the optimum mixture ratio for any propellant combination can only be determined by the optimisation of the total vehicle where mixture ratio is included as one of the variables. If we take as an example a stage that utilises an engine of almost zero weight, then the tankage dominates the structure and its minimisation is important and a mixture ratio giving minimum tank size commensurate with performance is required. If on the other hand the engine of the stage is very inefficient

Specific Impulse - Alkanes & RP-1 with LOX Oxidizer 380

360

340

320

300 Methane 280 Ethane 260 Propane Butane 240 RP-1 220

Theoretical Vacuum Specific Impulse, sec 200

180 0123456 Oxidizer/Fuel Mixture Ratio by Weight Figure 4.3: Theoretical vacuum specific impulses for light hydrocarbons and RP1 with LOX oxidiser. Chamber pressure is 10 MPa and other engine parameters are typical first stage values. Curves calculated using NASA-Glenn CEA program [Ref 4.15]

51

Chapter 4 Strategic Propellant Selection being very heavy then the minimisation of the tank size and mass becomes less important and the higher performance mixture ratio is of more importance. Consequently it is not possible to say which propellant combination is best by direct observation. What can be stated though is that, from the above data, on performance and density is that, from the above data on performance and density grounds propane would seem to be slightly more desirable than RP1 as a fuel when burnt with liquid oxygen. To determine whether Methane’s higher Isp would give it the overall performance edge would require a complete vehicle optimisation comparing the three fuels. Table 4.4: LOX/Propellant theoretical density specific impulse values as a function of mixture ratio. Green cells represent maximum Ivac while blue cells represent maximum Id

Mixture Theoretical vacuum density specific Ratio, O/F impulse, Id, sec.kg/litre by mass Methane Propane RP1 0.2 97.4 159.5 158.11 0.4 121.2 188.1 189.40 0.6 140.0 209.4 211.73 0.8 155.8 226.4 229.22 1.0 169.5 240.7 243.84 1.2 182.5 254.6 261.08 1.4 196.2 272.5 278.45 1.6 211.8 291.3 295.03 1.8 227.4 307.9 311.88 2.0 241.7 321.0 324.71 2.2 254.1 331.0 334.13 2.4 264.5 338.4 340.52 2.6 273.2 343.4 344.06 2.8 280.3 346.1 345.05 3.0 285.8 346.9 344.21 3.2 289.8 346.3 342.49 3.4 292.4 345.0 340.41 3.6 293.9 343.3 338.17 3.8 294.6 341.5 335.85 4.0 294.8 339.7 333.52 4.2 294.8 337.8 331.18 4.4 294.7 335.9 328.84 4.6 294.4 334.0 326.49 4.8 293.9 332.1 324.16 5.0 293.5 330.2 321.82 5.2 292.9 328.3 319.48 5.4 292.3 326.4 317.15 Ivac @ Max Id 342.9 349.0 335.1 r @ Max Id 4.2 3.0 2.8 Max Ivac 356.9 350.6 336.0 r @ Max Ivac 3.0 2.8 2.6

Case studies comparing Methane, Propane and RP1 as Launch Vehicle Fuels: While RP1 has been placed in the “Bad” category above it has been so for strategic reasons as opposed to performance reasons; however RP1 is still under consideration as a fuel for future launchers. The literature describes many studies describing the results of comparing

52

Chapter 4 Strategic Propellant Selection one propellant with another3 and a literature search reveals several papers that compare methane, propane and RP1. As far back as 1983, James A Martin [Ref 4.18] of NASA Langley Centre compared methane and propane as propellant for a single-stage to orbit vehicle (SSTO). The conclusion reached was that for the SSTO single-fuel engine under consideration propane was somewhat better than RP1 and considerably better than methane in reducing vehicle dry mass.

In 1983 a paper by Manski and Martin [Ref 4.19] again showed that for an SSTO LOX/hydrogen gave the lowest dry mass followed by LOX/propane then LOX/methane but did not compare RP1.

LOX/Fuel Density Impulse 400

350

300

250

200 LOx/Methane sec.kg/litre LOx/Propane 150 LOx/RP1

100 Theoretical Vacuum Density Impulse, Impulse, Density Vacuum Theoretical

50 0123456 Oxidizer/Fuel Mixture Ratio by Weight Figure 4.4: Theoretical vacuum density impulses for light hydrocarbons and RP1 with LOX oxidiser. Chamber pressure is 10MPa and other engine parameters are typical first stage values. Curves calculated using NASA-Glenn CEA program More recently in 2002, Burkhardt, Sippel, Herbertz and Klevanski carried out a study [Ref 4.20] intended to investigate the utility of replacing the Ariane 5 EAP solid boosters with liquid propellant boosters as an evolution of the Ariane 5 vehicle. In the study they investigated performance gains, reduced ecological impacts and costs possible resulting from the replacement boosters which were proposed to be reusable flyback units. Propellant candidates examined were LOX/LH2, LOX/Kerosene and LOX/Methane. Conclusions drawn were that a methane fuelled engine would be about 19% heavier than the equivalent RP1 fuelled engine and the Ariane-5 lift-off weight with methane fuelled boosters would be ~8300kg (about 1%) heavier than with RP1 fuelled boosters. The approximately 10s of increased specific impulse from methane was offset by the heavier engine, and the larger tankage required due to the lower density of methane. This also increased the aerodynamic drag on the vehicle. The increased dry mass of the reusable methane boosters was expected to cause a cost increase in the launch vehicle system. On

3 For example, AIAA Journal of Spacecraft and Rockets, AIAA Journal of Propulsion and Power, Journal of Hydrogen Energy, Proceedings of the AIAA/ASME/SAE/ASEE Joint Propulsion Conferences and Exhibits

53

Chapter 4 Strategic Propellant Selection the other hand, the authors state that if operational and maintenance costs for methane proved lower than for RP1 then methane could very well become competitive with RP1. The study did not examine expendable vehicles but it was expected that the conclusions would be similar. Other works in the literature [Ref 4.14] indicate that engine sooting and coking from the light hydrocarbons4 (e.g. methane, propane) is less than that from RP1 so that there are no serious engine design problems in this regard. Methane however has a considerably longer ignition delay than either RP1 or propane leading to a longer L* and hence engine length and weight From the above case studies we can conclude that, in agreement with the qualitative conclusions drawn from the specific impulse performance and specific density impulse figures of section 4.3.1.4 above, on performance and vehicle structural grounds, propane would appear to be the superior fuel followed by RP1 and methane in that order.

4.3.2 Advanced Launch Vehicle Fuels In addition to the fuels discussed above there are a number of other chemicals that could be utilised, some of which are even more energetic. Amongst these are cyclopropane, propyne, propene and ethylene. Cyclopropane: is more energetic than propane as the cyclic nature of the molecule means that strain has been added to the bonds increasing the heat of formation. It is believed that the Russian synthetic kerosene "sintin” was composed of molecules of the cycloparaffin row. Several seconds of Isp were gained through bond strain and several more by prechilling before loading propellant. The Sintin production plant appears to no longer exist probably due to the expense of production [Ref 4.3].

Methyl Acetylene, (Propyne): Acetylene (C2H2, Ethyne) burnt with liquid oxygen would provide a very high specific impulse. However, it is unstable at high pressures and the flame temperature is too high for effective chamber cooling. By methylating acetylene to produce methyl acetylene, (C3H4, Propyne), a much more tractable high performance fuel can be produced. With LOX as an oxidiser a specific impulse of 370s can be expected [Table 4.2]. Propyne has a relatively high density (s.g. 0.53), giving it a high density impulse figure. The freezing temperature of -102.7°C and the boiling point of -23.2°C cause fewer problems in storage than cryogenic fuels. It is also much less toxic than MMH (mono-methyl-hydrazine). A European space industry study by Valentian et al [Ref 4.21] comparing a number of possible propellant combinations, showed that propyne would be highly advantageous as a fuel for low earth orbit operations. It can also be used with non cryogenic oxidisers to give moderate specific impulse, e.g. with N2O4 Isp=338s. Drawbacks are that, unlike hydrazine, it is not a monopropellant and is relatively expensive to produce. Valentian et al. study future propellant use in boosters, manned and unmanned planetary missions and small spacecraft for the purposes of utilising “Green Propellants”. Their main rejection criteria are any carcinogenic and toxic properties. They carry out trade studies that include production and handling costs as well as the pure performance figures. They find that for large boosters N2O4/HC is the best option but if N2O4 is not accepted then LOX/LH2 is the best choice for upper stages if volume (density) is not a problem, otherwise LOX/HC for upper and lower stages. In Australia N2O4 is probably not acceptable for environmental and safety reasons. Likewise N2O would also be unacceptable as an oxidiser. While N2O is non-toxic and only narcotic in high concentrations any reactions could potentially form toxic compounds of nitrogen and the

4 Light hydrocarbons abbreviated as LHC, hydrocarbons as HC

54

Chapter 4 Strategic Propellant Selection safety issue would again arise. LHCs are also non toxic and are “soft cryogens”. Like oxygen they can be stored indefinitely with super-insulated tanks utilising existing cryo-refrigerators. On these grounds too they would again seem to be the all round choice for large boosters.

For high Isp, Valentian et al also discuss the use of propene, propyne, ethylene and ethers in boosters. Of these LOX/propyne has the highest Isp of 370s (Valentian et al quote this figure at a chamber pressure of 1.03MPa and ε =40), 5 sec greater than LOX/methane. The cost of production, compared to the LHCs, as well as the potential toxicity and possible instabilities at high engine temperatures and pressures would seem to rule these out in any nascent space launch industry such as in Australia.

4.4 PROPELLANT AVAILABILITY IN AUSTRALIA

4.4.1 Liquid Oxygen as Oxidiser of Choice

There are only a limited number of choices of oxidiser: LOX, H2O2, N2O, N2O4 and the nitric acids. For launch vehicle boosters where there is a large volume of propellant, 50 - 300 tonnes, it is desirable to rule out the nitrogen containing compounds for the toxicity of their reaction products. This leaves only LOX and H2O2. Peroxide has the advantage of being a liquid at room temperature with a specific gravity of about 1.4 but it returns a relatively low specific impulse with most fuels although Dunn Engineering [Ref 4.8] claims it to be superior to LOX/RP1 when used with propylene. This combination should be ruled out for launch vehicle use in Australia because of the possible environmental and political problems of manufacturing, transporting and storing these chemicals in large quantities. It is also possible that the environmental movement would find the use of propylene an excuse to oppose the project. In the past they opposed the launching of Russian Proton launch vehicles from the proposed Cape York International Spaceport. Their worry was that the residual hydrazine of spent stages would pollute the waters of the Great Barrier Reef. In addition H2O2 can be unstable and difficult to handle and is not manufactured in Australia. LOX has been used extensively in launch vehicles; its properties are well known, it’s relatively easy to handle and non-toxic as well as being low-cost and plentiful, of common industrial use and is produced in Australia in quantity. Its only drawback is that it is cryogenic requiring chill down time for loading into the launch vehicle. It can be stored indefinitely using common technology.

4.4.2 Launch Vehicle Fuels Alcohols: In Australia Manildra is a supplier for Ethanol. The product is 99.8% pure at a cost of approx $1.00/litre. They produce 100Ml/yr and state that they could also be produce Propanol and Butanol biologically but no production project is proposed [Ref 4.22]. CSR are the second source supplier for Ethanol. They manufacture the raw feedstock at their sugar mill in Serena, Queensland. This is then treated to reach the required purity at the plant at Yarraville in Melbourne. Production is 55-60Ml/yr, Sulphur content is <5ppm. Grades are 100SG <0.2% water (typically 0.1-0.15%) & 100HG <0.05% water. CSR import n-Propanol from the USA, typically 200 tonnes every 3 months but don’t deal in Butanol [Ref 4.23]. In 1943 CSR-Chemicals (then Robert Corbett Pty Ltd) adapted the Weizmann process for the production of Butanol from Sugar to produce Dibutyl-phthalate for use as an insect repellent [Ref 4.24]. CSR-chemicals have since been sold to Orica Pty Ltd. and discussion

55

Chapter 4 Strategic Propellant Selection with their chemical supplier subsidiary, Spectrum Distributors, reveal that any facilities that could be utilised for the production of Butanol have been closed down and any n- or iso- Butanol used in Australia is currently sourced from BASF [Ref 4.25].

Butanol is considerably better than Ethanol in terms of Isp and density for launch vehicle use (see table 4.2) and while the Ethanol economy is gaining momentum to replace or supplement petroleum fuels in motor vehicles, the bio-butanol economy also has its proponents. For example, DuPont and BP have formed a consortium to investigate the production of bio-butanol from waste agricultural materials as feedstocks [Ref 4.26]. The company “Greenbiologics” in the United Kingdom are also proposing production of bio-butanol by a patented hydrolysis technique [Ref 4.27]. They propose “bolting-on” a bio-butanol plant to current ethanol production facilities. While there is currently no Butanol production in Australia it would appear that there will shortly be a bio-butanol economy that can be sourced for launch vehicle propellant either domestically or internationally.

Light Hydrocarbons: The story of propellant availability in Australia is basically that of the petrochemical industry that has arisen in the last decade. Methane has been available in Australia for many years as it is the major component of natural gas. However, it has recently become available as a purified liquid. The famous North-West Shelf offshore oil and gas are operated by Woodside and there are five complete processing trains for LNG (Liquefied Natural Gas) and LPG (Liquid Petroleum Gas) located at Karratha and the Burrup peninsula in the Pilbara. This gas is mainly destined for export to China, South-East Asia and Japan. However some of the gas is utilised for domestic supply. Woodside supplies its natural gas to Wesfarmers energy division via the Dampier-Bunbury pipeline. Wesfarmers separate the methane from the other light hydrocarbons (mainly ethane, propane and butane) at their Kwinana plant south of Perth. For several years Wesfarmers have operated a pilot plant [Ref 4.28] to liquefy the separated Methane which is then marketed by Kleenheat as LNG [Ref 4.29] and the Propane as LPG [Ref 4.30]. The pilot plant also located at Kwinana south of Perth is only capable of supplying 6 tonnes per day of LNG; however, a production plant currently under construction will be capable of supplying 175 tonnes per day. The plant was scheduled to be commissioned at the end of the first quarter of 2008. Part of the production capability (60 tonnes a day) is already earmarked to be supplied to AngloGold Ashanti’s Sunrise Dam and Barrick’s Darlot gold mines to run facilities there. The LNG is shipped in trucks capable of carrying 30 tonnes each. Wesfarmers also intend to extend their distribution network to other states and has also successfully concluded long-term gas supply and pipeline access arrangements with Santos Ltd and Dampier-Bunbury Pipeline respectively. The LNG, as supplied, is low in sulphur (<1ppm) which has been removed at the LPG stage. The main pollutant of the LNG is nitrogen which is at the 2-3% level. The nitrogen can be removed by Wesfarmers/Kleenheat at Kwinana for a price, however, discussions with Wesfarmers representatives have indicated that it would be possible to remove the nitrogen at the launch site by flashing, i.e. careful temperature control to just below the boiling point of methane and release of the pressure. The Nitrogen will then boil off. Wesfarmers have indicated that the price of the LNG is about $0.40 per litre. At a specific gravity of methane of 0.43 and allowing for the nitrogen content that equates to $0.96/kg. Wesfarmers process the LPG further to extract Propane which can be produced at 98.4% purity5 with approximately 1% Ethane and 0.45% Butane remaining. The sulphur content

5 In the UK at least, Intergas supply Propane as R290 refrigerant in two grades. The highest purity is N2.5 grade of 99.5% purity with volume impurities at: Nitrogen <400ppm, Oxygen <100ppm, Carbon dioxide <100ppm, Other Hydrocarbons <4500, Moisture <10ppm [Ref 4.31]

56

Chapter 4 Strategic Propellant Selection is approximately 3ppm and exists as carbonyl sulphide (O=C=S) with no Hydrogen Sulphide (H2S). This is considerably lower than the 50ppm currently allowed in the RP1 specification. In addition the Propane produced contains no Olefins6, triple bond, cyclic or aromatic compounds, hence is free from the possibility of polymerisation7. Other non-refinery sources of propane are available in Australia. At the Cooper basin in South Australia, Santos produce very low sulphur content light, sweet premium crude oil suited to the production of high quality transport fuels. Pressurised propane and automix are produced at Port Bonython for loading into road tankers for the wholesale market. Propane delivered through this terminal is a high quality product with current production running at approximately 170,000 tonnes per annum Propane and 100,000 tonnes per annum Butane. Specifications of the propane supplied by Santos are:

Table 4.5: Cooper Basin Propane - Quality Specifications

Components Liquid volume % Guaranteed Typical Ethane 1.6 Propane 95.0 97.1 Butane 2.5 1.3 Pentanes Plus N/A <0.01 Volatile Sulphur 30.0 max <1.0 (ppm/wt)

Hydrogen, Ammonia and other petrochemical derivatives: The world's largest greenfield ammonia plant, costing US$575 million, located near Karratha on the Burrup peninsula in Western Australia was opened in April 2006 [Ref 4.32]. It employs about 100 people at full production with an annual production capacity of 776,000 tonnes of liquid ammonia, with production of hydrogen, methanol, ethylene and other side products.

4.4.3 Pressurants Helium: Where pressurants are required in spacecraft, helium is preferred for its light weight. It’s considered a very strategic material and is in short supply worldwide with the major supplier currently being the USA. In Australia, Central Petroleum started drilling for helium southwest of Alice Springs at Mount Kitty in 2007. High readings of Helium (6%) were found during drilling in the early 1990s and the seismic results also looked promising. Central Petroleum have said that even if their drilling at Mt Kitty does not yield Helium “they have plenty of other leads” and they expect someone to find helium in Australia within the next five years [Ref 4.33]. Nitrogen: For pressurants where weight or solubility is not a problem, nitrogen is available in large quantities from the fractional distillation of air and is available in quantity from gas suppliers in Australia.

6 Olefin = an unsaturated chemical compound containing at least one carbon-to-carbon double bond, e.g. Alkenes 7 Olefins and higher bond compounds are found in Propane produced from refineries.

57

Chapter 4 Strategic Propellant Selection

4.4.4 Spacecraft Propellants The smaller quantities of propellant involved in the spacecraft relative to the launch vehicle make the choice of propellants less critical from the environmental, political and operational standpoint although there would still be some limitations on their use. However, the requirement for on-orbit storage of propellant introduces new problems. N2O4, N2O, H2O2 are suitable as oxidisers with Propyne added to the list of fuels. Again see Valentian et al [Ref 4.21] for a full discussion on this topic.

4.5 CONCLUSIONS Consequent to the above discussion it is possible to narrow down the choice of propellant combinations for use in an Australian space launch vehicle. Propellants containing fluorine, chlorine and boron have already been eliminated mainly on environmental grounds but also as being too expensive and toxic [Ref 4.2, 4.21]. The propellants using less common propellant chemicals such as propyne or propylene can also be eliminated on cost, environmental and political grounds. The remaining propellant combinations are shown in the trades in table 4.6 along with the N2O4/Hydrazine combination that is also to be discounted on environmental and toxicity grounds. To summarise then, the remaining combinations are:

• LOX/LH2 • LOX/HC • LOX/Alcohols

• H2O2/Kerosene Where the HC fuels are RP1 and the paraffins, Methane to Pentane (say) and the alcohols are the n- (or 1-,) alcohols starting with Methanol, Ethanol and extending to n-Pentanol (say).

LOX/LH2 is preferred for upper stages because of its high performance but is contraindicated for first stages because of the low density of LH2 requiring large tankage. Of the Alcohols, Ethanol is preferred for availability, cost and ease of handling. Butanol is preferred for availability, performance and ease of handling. Of the Hydrocarbons, Methane is preferred for performance, cost and availability but disadvantaged by its low density especially for first stages. Propane is preferred for its still high performance, cost, availability and temperature compatibility with Liquid Oxygen, i.e. its operational, handling and storage properties. RP1 is preferred for its handling, storability and its historical technological data base, but disadvantaged in the Australian environment by the requirement to build or modify a refinery for its production. Studies have shown that on performance grounds there is little to split Methane, Propane and RP1.

H2O2/Kerosene is viable for small vehicles as exemplified by Black Arrow but its relatively low performance would rule it out for large vehicles because of stage size and stability issues unless there were very good operational reasons for using it. Three alternative layouts for oxidiser and fuel tanks are shown in Figure 4.5. The tandem tank layout is utilised when the propellants are of completely differing temperature regimes such as LOX/RP1 while the common bulkhead layout is used when intertank insulation can be used to prevent gross heat flow between propellants of nearly identical temperatures such as LOX/Liquid-Methane. This layout has the advantage of smaller size and lower mass than the tandem tank layout.

58

Chapter 4 Strategic Propellant Selection Australia Australia Availability In materials LOX Available feedstock RP1 New available. refinery needed Available. Available. use Common materials Available use Common materials Available use Common materials Available. Available. use Common Not Available Not Available Not Available Cost Low Low Moderate Moderate Low High High Low High High Reliability Reliability Good Good No operational No operational yet. engines Good Potentially Good No operational yet. engines Potentially Good Excellent Excellent Hypergolic Good Good Good to Excellent

insulated at at insulated 8 Storability Storability H 3 LOX Limited LOX Limited Limited Limited LOX Temps Temps LOX Limited Limited Limited LOX C Good Good LOX Limited LOX Limited Good

Safety firehazard Explosion Explosion 2 2 O 2 hazard hazard LH LOX fire hazard hazard fire LOX LOX fire hazard hazard fire LOX LOX fire hazard hazard fire LOX LOX fire hazard hazard fire LOX Highly Toxic, Protective suits required LOX firehazard H Protective suits required Environmental Environmental Considerations Exhaust is is Exhaust Hydrocarbons is Exhaust is Exhaust is Exhaust diluted is Exhaust Hydrocarbons Exhaust is steam steam is Exhaust Hydrocarbons Highly toxic Highly toxic Hydrocarbons Hydrocarbons Hydrocarbons – Trade Table for Propellant – Trade Combinations Operational Operational Table 4.6 required required Considerations required required Cryogenic Cryogenic dewars Ground required Cryogenic dewars Ground required Cryogenic Cryogenic dewars Ground LOX Cryogenic LOX Cryogenic Ground dewars LOX Cryogenic Complex due due Complex due Complex issues Safety Ground dewars dewars Ground required Safety issuesSafety Very Low Very Density 2 large tanks tanks large Moderate CH4 Low Requires large tanks LH Moderate Requires Moderate Moderate Moderate Moderate High High High High Very High Very High High High High Moderate Moderate Performance Performance /Hydrazine(s) /Kerosine /Kerosine 4 2 Propellant O O 2 2 LOX/RP1 LOX/RP1 LOX/Methane LOX/Methane LOX/LH2 LOX/LH2 LOX/Propane N LOX/Alcohol(s) LOX/Alcohol(s) H

59

Chapter 4 Strategic Propellant Selection

Figure 4.5: Alternative propellant tank layouts. a) Tandem tank layout for dissimilar propellants. b) Common bulkhead tanks for propellants with similar temperature properties. Intertank insulation minimises heat flow between tanks. c) Tank-in tank layout for propellants benefiting from the insulating properties of the outer tank propellants. This would be an ideal layout for LOX/Propane. Diagrams adapted from Huzel and Huang [Ref 4.34]

60

Chapter 4 Strategic Propellant Selection

The third tank layout, the tank-in-tank system, is used where propellant temperatures are identical or of no importance for example, H2O2/Kerosene. This layout is however also of considerable importance for the LOX/Liquid-Propane propellant combinations. If the outer oxidiser tank is filled with liquid oxygen, it will act as an insulator for the inner tank which is now in an environment at the LOX temperature. Liquid propane stored in the inner tank will be at a greatly increased density than at its boiling point and its vapour pressure will be almost negligible8 so that boil off during countdown or holding on the pad will be minimised. The only boil off will be of the LOX. It is assumed that the tanks have been made sufficiently well that there is no leakage between them! This layout gives a weight advantage as no insulation is required between the tanks and structural strengthening placed between the tanks can be used to minimise stage weight.

8 At LOX temperature liquid propane vapour pressure is 0.1mm Hg whereas near its boiling point the vapour pressure is about 5 Bar.

61 Chapter 5 Launch Vehicle Stack Optimisation by Evolutionary Algorithm

5.1 STACK MODEL As the first step towards a more detailed vehicle model, the Gross Lift-Off Weight (GLOW) of the vehicle was required to be minimised on the initial assumption that the lowest-cost vehicle would correspond to the lightest-weight vehicle. Evolutionary optimisation techniques were used to choose the masses for the three stages that would minimise the GLOW. A number of other parameters, as described below, were assumed constant and were not optimised. It was initially desired to model a clean-sheet design for a three-stage launch vehicle with payload and much of the design being equivalent to that of the Ariane 44L (AR44L) vehicle. While the AR44L design is a three and a half-stage vehicle, due to four liquid fuel strap-on boosters being staged in parallel with the main first stage of the vehicle, the baseline Hypothetical Launch Vehicle (HLV) design was to be a straight three-stage vehicle (Figure 3.1), as discussed in Chapter 3. The questions to be answered were: how would the weight of this design compare to that of the AR44L; how would the evolutionary methods perform when determining a minimum GLOW and how sensitive would be the results?. In addition, the model had to be able to be used to design other proposed launch vehicles and move away from the AR44L design parameters.

5.1.1 Method In order to obtain realistic mass and ∆v values for the model, the AR44L launch vehicle was used as a baseline against which to compare results from the model. Analysis of the performance capability of the AR44L was carried out by applying the rocket equation to each stage to obtain the ideal velocity of the entire vehicle1. The ∆v of the AR44L was determined to be 11945.5 m/sec and the ∆v required from the hypothetical three-stage HLV was to be the same. While the masses quoted for the AR44L components vary from reference to reference, those used were the best available from easily obtainable sources, primarily the Ariane User’s Manual [Ref 5.1]. Specific impulse performance and structure factor data for each stage were from the same source. Table 5.1: Payload and equipment masses used for both AR44L and HLV models Item Mass, kg Fairing 900 Payload 4568 Adapter 200 Vehicle Equipment Bay 530

Identical payload and accommodation masses to those in the AR44L vehicle were used for the HLV model and are shown in Table 5.1. The HLV was initially modelled on an MS- Excel spreadsheet program (STAGEX) and, in parallel, a FORTRAN program (ROKOPT) was written to duplicate its results. ROKOPT was to be the basis of a full launch vehicle model and was used here to test the evolutionary optimisation methods to be used later on a more detailed model.

1 An explanation of the method for calculating the ideal velocity is detailed in Chapter 6

62

Chapter 5 Stack Optimisation

5.1.2 Spreadsheet Method The initial methodology involved displaying the HLV stack mass breakdown on the STAGEX spreadsheet and carrying out an optimisation to maximise the ∆v delivered by the stack. The propellant specific impulses and the stage structure factors were assumed to be identical to those of the AR44L vehicle. Implicit in this scenario is the assumption that the propellant combinations used in the two vehicles are also identical. In reality, identical structure factors are not likely to occur because of the greatly differing sizes of the stages of the two vehicles. The Ariane payload fairing is jettisoned during the second-stage burn at a height that depends on the aerothermal flux that can be tolerated by the payload. This can make a difference of up to 14kg in the payload and a variation of 14 seconds in the mission time for the fairing jettison. In the HLV, the fairing is considered to be jettisoned either just before or at the second-stage burnout. The initial stage masses, including the propellant, chosen by any of a number of methods (including guessing) were entered into the spreadsheet and modified until the desired ∆v was obtained. The initial stack was modified by entering mass exchange values into dedicated cells to investigate the effect on the ∆v of moving the mass from one stage to another. Once the ∆v was maximised, the whole vehicle was re-sized to, once again, give the target ∆v. The process was repeated until the minimum GLOW was achieved. This method illustrates the optimisation of a launch vehicle stack with a single objective function (GLOW) and a single constraint (∆v) to be observed. The results of the manual spreadsheet optimisation gave a GLOW of 356,152 kg for a target ∆v of 11945.5 m/sec, about 124 tonnes lighter than the 480,048kg AR44L weight. The resultant stage masses are shown in Table 5.2. Table 5.2: Optimal stage masses for HLV as determined by STAGEX HLV Stage Mass, kg Stage 3 42630 2 86104 1 221220 GLOW 356152 ∆v 11945.5 m/sec

The result of the STAGEX optimisation shows that, as a consequence of effectively reducing the size of the first stage by eliminating the four liquid strap-on boosters, the HLV vehicle requires its two upper stages to be considerably larger than those of the AR44L. The first stage is about 53% of the combined mass of the AR44L’s first stage and strap-on boosters. 5.1.3 Velocity Budget The equations for launch vehicle sizing are well-known and can be found in many texts. For example, White [Ref 5.2] develops the equations for single, multiple and infinite numbers of stages with like and unlike stage parameters. There is, however, no analytical method which allows the equations describing a multi-stage launch vehicle to be solved for the optimum mass ratios of each stage if the effects of drag and gravity are considered [Ref 5.3]. Vehicle design is dependent on the ∆v to be achieved, which is partly dependent on the trajectory flown which, in turn, is dependent on the vehicle design. Therefore, there is no closed form solution to the problem. However, if the velocity requirement (or budget) is known, a preliminary launch vehicle design can be produced to provide the required ∆v. Approximations of the velocity budget to be flown can be made, including allowances for thrust-atmospheric loss, drag loss, gravity loss, earth rotation gain and orbital velocity.

63

Chapter 5 Stack Optimisation

5.1.4 Detailed Launch Vehicle Model The STAGEX spreadsheet described above was developed to assist with the estimation of the velocity budget. While the estimates are not exact, they are sufficient to allow an estimate of the stage sizes to be made for use as inputs to a more detailed optimising computer model which will eventually include integration of the trajectory to provide guidance parameters and the flight profile. Thrust-atmospheric loss effects, drag, the effects of the mixture ratio on the delivered specific impulse, the propellant density and, hence, the structure mass will also be included. These further developments of the model would include many more variables to be optimised than were in the simple stack model. It would then not be possible to optimise the variables manually as was done with the STAGEX spreadsheet. Thus, an automatic method must be found. In order to test optimisation methods for the full computer model (ROKOPT), a project was set up to run on a Pentium-4 desktop computer under the Windows-XP operating system. The development environment used was MS-Visual Studio with an integrated Compaq Visual Fortran 6.6C compiler. Mixed language projects are supported in this environment so that, although the main code was written in Fortran-77 with Fortran-90/95 extension, C++ subroutines could also be automatically compiled and linked within the project. The initial part of the project was to calculate the stack GLOW and ∆v using the data from the HLV vehicle, as approximated by the STAGEX spreadsheet. Optimisation code was then added to choose stage masses for minimising the GLOW and constraining the HLV’s ∆v to the AR44L’s ideal velocity.

5.2 EVOLUTIONARY METHOD Optimisation methods rely on the generation of an initial random population of solutions to the problem in which the variables are randomly chosen from a given range near the optimum values. The function is then evaluated for each of the sets of generated variables (designs). Following this, the algorithm generates a child population (set of designs) and the better solutions among the set of parents and the child population which survives to the next generation are observed. After a number of generations, the population will converge towards a minimum objective function with the constraints being observed. In the case of a launch vehicle stack, the variables are the stage masses with their specific impulses and structure factors as constants. The objective function is the minimum GLOW and the constraints to be observed are the upper and lower values of the allowed ∆v velocity range. Populations of random values for the stage masses lying within the allowed mass ranges were generated and then the GLOW was evaluated and the ∆v capability determined for each set. The ∆v values may not be within the allowed tolerance for the initial generation and can take on any values corresponding to those given by the randomly selected values of the variables. As the generations go by, the evolution rules soon move the variable sets into regions where the ∆v constraint is met. The evolutionary optimisation method used in this study is described in detail in Ray and Sarker [Ref 5.4]. Their evolutionary algorithm (EA) is a variant of NSGA-II [Refs 5.5, 5.6] with a modified method of population reduction which insists on maintaining the diversity of solutions in both the objective function and the variable space. The method is computationally more expensive than NSGA-II but maintains the diversity of the variable space more effectively than does NSGA-II. A description of the NSGA-II method is given below as described by Ray

64

Chapter 5 Stack Optimisation

Non-dominated Sorting Genetic Algorithm (NSGA-II) is one of the most popular population based optimisation algorithms that have been successfully used in a number of real life applications. The main steps of NSGA-II are outlined in Algorithm 1. The algorithm starts with an initial population (P1) of N candidate solutions initialised by random sampling from the design (variable) space. Each candidate solution of the population is evaluated to yield the corresponding values of the objective and the constraint functions. Based on the objective and constraint function values, the candidate solutions are ranked. The next few steps (lines 5-8) are repeated for NG generations. An offspring population, Ci is created using recombination operation from the current (or parent) population Pi−1. The new offspring solutions are evaluated and the combined set of parent and offspring solutions is ranked. Based on the ranks, the best solutions from the parent population Pi−1 and the offspring population Ci are retained to form the population for the next generation Pi.

Algorithm 1: Non-dominated Sorting Genetic Algorithm-II (NSGA-II)

Require: NG > 1; {Number of Generations} 1: Initialize (P1); {Create an initial population of solutions} 2: Evaluate (P1); 3: Rank (P1); {Assign ranks to each solution} 4: for i = 2 to NG do 5: Ci = Evolve (Pi-1); {Create child solutions from parents of previous generation} 6: Evaluate Ci; {Compute the performance of the child solutions} 7: Rank (Pi-1+Ci); {Assign ranks to each solution} 8: Pi = Reduce (Pi-1+Ci); {Identify parents for the next generation} 9: end for

Initialisation: All the individuals in the population are initialized by random sampling. A value for each design variable is sampled uniformly between the lower and the upper bound for the variable as given in the equation below:

xix = ii+− U[0,1] ( xx i ); 1 ≤ i ≤ n

where xi denotes the initialized variable xi and xi are lower and upper bounds for the variable, and U[0, 1] is an uniform random number lying between 0 and 1.

Evaluation: For each solution in the population, the values of the objective and the constraint functions are evaluated using appropriate simulation or analysis. The fitness of a solution is calculated based on the objective and the constraints values as follows: 1. For a feasible solution, the fitness corresponds to the objective value(s). 2. For an infeasible solution, the fitness corresponds to the value of the largest constraint violation. Evolution: In NSGA-II, an offspring population is evolved from the current population using selection, crossover and mutation operations. The details of each are provided below. Selection: In NSGA-II, two parents are selected to create two offspring. The selection of each of these parents is based on a binary tournament. The process of binary tournament is illustrated below to identify a parent and the same needs to be repeated to identify another mating parent.

65

Chapter 5 Stack Optimisation

a) Among two potential parents x1 and x2, if x1 is feasible and x2 is infeasible, x1 is selected as the parent and vice versa.

b) If both x1 and x2 are infeasible, the one for which the value of the maximum constraint violation is smaller is selected.

c) If both x1 and x2 are feasible and x1 dominates x2, x1 is selected and vice versa.

d) If both x1 and x2 are feasible and neither dominates the other, one of x1 and x2 is selected at random. Crossover: The crossover operation is performed between two parents identified above using simulated binary crossover (SBX) [Ref 5.7]. Two offspring solutions y1 and y2 are created from parents x1 and x2 by operating on one variable at a time as shown below. 1 1 2 yi =0.5[(1+βqi) xi + (1 - βqi) xi ] 2 1 2 yi =0.5[(1 - βqi) xi + (1 + βqi) xi ] where βqi is calculated as,

1  +1  (2uifu )ηc , ≤ 0.5,  ii β = 1 qi  1 +1  ηc (), if ui ≥ 0.5.  21()− ui and where ui is the uniform random number in the range [0, 1) and ηc is the user defined parameter, Distribution Index for Crossover. Probability of crossover (Pc) determines how often the crossover operation is performed. Mutation: Polynomial mutation operator Ref 5.8 is used for mutation. In the mutation operation, the value of one or more variables is randomly perturbed as given in the equation below.

yxii=+()xi −xi δ i where δi is calculated as, 1  −1  (2rifr )(1)ηm + , ≤ 0.5, δ = ii i  1  (1)ηm + 1−− (2(1rifrii )) , ≥ 0.5. and where ri is the uniform random number in the range [0, 1) and ηm is the user defined parameter, Distribution Index for Mutation. The numbers of solutions undergoing mutation operation are determined by probability of mutation (Pm). Ranking: Individual solutions in a population are ranked based on their fitness value. Feasible solutions are considered better than infeasible solutions and are ranked higher. Feasible and infeasible solutions are ranked separately. For single objective optimization feasible solutions are sorted based on the objective value while for multi-objective optimisation the solutions are ranked based on non-dominance. The NSGA-II uses non-dominated sorting and crowding distance sorting procedure [Ref 5.6] to rank feasible solutions with multiple objectives. In non-dominated sorting the solutions are arranged in multiple non-dominated fronts. In each non-dominated front, the solutions are non-dominated, whereas the solutions in one front dominate the solutions

66

Chapter 5 Stack Optimisation from the other front. Within a non-dominated front, the solutions are ranked based on a diversity measure, crowding distance [Ref 5.6]. For infeasible solutions the fitness corresponds to the maximum constraint violation. If more than one constraint is violated for a solution, the largest constraint violation value is used for the maximum constraint violation. Infeasible solutions are sorted in the increasing order of maximum constraint violation value. Reduction: The reduction process is used to retain N best solutions from a set of 2N solutions (parent and offspring populations) for the next generation. It uses the fitness values or ranks obtained through the above explained ranking procedure. 1. If there are more than N feasible solutions, • N feasible solutions are selected in the order of non-dominated fronts and decreasing crowding distance in each front. 2. If the feasible solutions are less than or equal to N, • all the feasible solutions are selected in the order of non-dominated fronts and decreasing crowding distance in each front, and • the remaining solutions are selected from infeasible solutions in the order of minimum value of maximum constraint violation.

5.3 THE CALCULATIONS The problem was to test the optimisation algorithm to determine the number of members in the population, and the number of generations, required for the evolutionary computational method in order to obtain a minimum GLOW with the stack ∆v capability (ideal velocity) constrained to 11945.5 m/sec. The results of the calculation of the stack mass and the ideal ∆v delivered were coded to include the payload configuration of Table 5.1 and the fairing drop at the second-stage burnout. While the ∆v target represents one physical constraint, the minimisation routine was given two constraints, viz., ∆v +ε and ∆v -ε, where ε is the residual allowed in the ∆v at convergence. The routines were, therefore, working with three variables (viz., the three stage masses), one objective function to be minimised (GLOW) and two constraints to be observed. An acceptable residual, the ∆v constraint, ε, of 0.1 m/sec was deemed to be possible as it was found that the population very rapidly observed the constraint during the optimisation with the minimum GLOW taking longer to be found. As the minimum lies on the constraint boundary and because the objective function is very flat near the minimum a tight tolerance on the constraint is required to enable the minimum to be distinguished.

Table 5.3: Stage mass ranges for the HLV as input to ROKOPT HLV Stage Mass Range, kg Stage 3 0 – 60000 2 0 – 200000 1 0 – 400000 GLOW To be minimised Delta-V 11945.5 m/sec The initial stage size estimates were derived from STAGEX and the ranges used allowed for a large spread of input solutions to the problem (Table 5.3).

67

Chapter 5 Stack Optimisation

Table 5.4: Values of evolutionary parameters used to evaluate the minimum GLOW Parameter Values No of solutions 100 Max number of generations 50 Random Seed 19 off Probability of crossover 0.90, 0.92, 0.94 Probability of mutation 0.05, 0.07, 0.09 Distribution index crossover 10, 15, 20 Distribution index mutation 10, 15, 20

The random number generator used to generate the random solutions for this method was RAN0, according to the method of Park and Miller described in reference 5.9

5.3.1 Computer Runs A run of 1539 evolutions was carried out using all the combinations of the values of the parameters shown in Table 5.4. The number of input solutions generated was 100 although it is normally estimated that 10N should be used [e.g. Ref 5.10] where N is the number of variables (viz., the three stage masses). Table 5.5: Stage masses for the HLV as determined by ROKOPT Vehicle 1st stage 2nd stage 3rd stage GLOW, Statistics mass, kg mass, kg mass, kg kg Maximum GLOW 147858 172565 44629 371250 Minimum GLOW 221113 86088 42607 356006 Spread of values -73255 86477 2022 15244 Percentage spread 33.1% 100.4% 4.7% 4.3% Mean solution 219863 89689 42036 357785 Median solution 220656 87915 41556 356325 STAGEX solution 221220 86104 42630 356152 STAGEX residuals 107 -16 24 146 Fifty full generations (parent and child) were used but, while the population converged after 42 generations, the constraint was observed after 36 and took a total elapsed time of 4976.1 seconds on the Pentium-4 computer.

GLOW Distribution 374 372 370 368 366 364 362 360 GLOW, kg x 1000 358 356 354 0 200 400 600 800 1000 1200 1400 1600 1800 Population Member

Figure 5.1: Distribution of evolved GLOW population

68

Chapter 5 Stack Optimisation

Table 5.5 shows the maximum and minimum evolved GLOW solutions found The minimum vehicle is now 146kg lighter than the solution found by STAGEX. Figure 5.1 shows that the distribution of the evolved solutions is not linear but that some members are at extremely high GLOW values while most are near the minimum GLOW value. However, there is a smooth progression of the population to higher values of the GLOW. Approximately 1100 solutions accurately observed the velocity constraint with most of the rest of the 1539 having residual of less than 1.0 m/sec while a small number had residuals of up to 5.0 m/sec. When the residual is plotted as a function of the GLOW (Figure 5.2), a front is visible in which the residual approaches zero as the GLOW approaches the minimum.

Velocity Residual v GLOW 5.0

4.0

3.0

2.0

1.0 Velocity Residual, m/sec Residual, Velocity

0.0 354 356 358 360 362 364 366 368 370 372 374 GLOW, kg x 1000

Ideal Velocity Distribution 11951

11950

11949

11948

11947 Ideal velocity, m/sec 11946

11945 0 200 400 600 800 1000 1200 1400 1600 1800 Population Member

Ideal Velocity Residual Distribution 5.0

4.0

3.0

2.0

1.0 Velocity Residual, m/sec Residual, Velocity

0.0 0 200 400 600 800 1000 1200 1400 1600 1800 Population Member

Figure 5.2: Distributions of ideal velocity and ideal velocity residual

69

Chapter 5 Stack Optimisation

The spreads of the evolved stage masses are large but this is due to the high GLOW solutions in the population of 1539 runs. The last line in Table 5.5 presents the difference between the STAGEX solution and the minimum evolved solution found by ROKOPT. Confirmation that the runs reached a minimum optimal solution can be obtained by examining Figure 5.3. It can be seen that, at the lower GLOW values the spread reduces to almost zero.

Third Stage Mass Solutions 55

50

45

kg x 1000 40

35 Third stage MassSolutions

30 355 360 365 370 375 GLOW Solutions, kg x 1000

Second Stage Mass Solutions 180

160

140

120

100 kg x 1000

80

Second Stage MassSolutions 60

40 355 360 365 370 375 GLOW Solutions, kg x 1000

First Stage Mass Solutions 300

280

260

240

220

kg x 1000 200

180 First Stage MassSolutions 160

140 355 360 365 370 375 GLOW Solution, kg x 1000

Figure 5.3: Evolved stage mass solutions against their evolved GLOW solutions The STAGEX operator gave up trying to find the absolute minimum GLOW at about the 0.1 tonne level. From the improvement in the GLOW, it can now be assumed that the evolutionary methods used by ROKOPT came close to finding the absolute minimum within that level.

70

Chapter 5 Stack Optimisation

Because of the difficulty of finding the absolute minimum, it would appear that the objective function is flat near the optimum. This could be useful if, in the detailed design of one stage, the mass overruns the desired figure, but only if the overrun can be compensated for in the other stages. It is also apparent from Figure 5.3 that the solutions are bounded by an envelope. The exact nature of this envelope is unclear in this visualisation but its nature will be discussed later. However this does not mean that either the STAGEX or ROKOPT solutions reached the absolute local minimum value of the GLOW. The optimisation using STAGEX became difficult around the minimum as the solution wandered around the basin of the local minimum. It is not obvious when operating the spreadsheet which way to move the solution to reach the absolute minimum. On the other hand, the evolutionary method will eventually choose a minimum and, given enough time, will automatically perform better than STAGEX

5.3.2 Specifying the Constraints In the ROKOPT run discussed above, the single constraint was the requirement for the vehicle’s ideal velocity to be equal to that of the Ariane 44L, viz., the calculated value of 11945.5 m/sec. In order to achieve this physical constraint, the algorithm was given two numerical constraints, an upper and a lower value, to limit the velocity to the band between the two values. Subsequently, another method was tried. It was reasoned that, because the velocity achieved (with a given payload) was dependent on the size of the launch vehicle, it would be possible to constrain the velocity with one constraint only, namely, that the velocity would be greater than the required value. With the objective function being the minimisation of the launch vehicle lift-off weight, the velocity would automatically be decreased until it equaled the constraint value. Test runs showed that this indeed produced the required result. However, it was realised that this would only be true if the constraint variable was a direct function of the objective function. Otherwise, a range for the value unless the exact value of the constraint variable was only required to be greater or less than a given value (inequality). Imposing an equality requirement, such that the constraint variable should be exactly equal to the required value, would be too stringent a test as all members of the initial population would then be found to be infeasible.

5.3.3 Gradient Methods to Supplement the Evolutionary Methods At the bottom of the basin of the minimum of the objective function, it is difficult to find the absolute minimum. This was evident when using the STAGEX spreadsheet as the operator couldn’t determine the best direction in which to move to find an improvement in the objective. A commercial solver add-in routine for STAGEX was obtained to automatically optimise the stack model. The solver routine used an evolutionary method to find the approximate solution and then utilised a gradient method to find the minimum of the basin. The results obtained using this software agreed exactly with the results obtained using the ROKOPT program, thus verifying the ROKOPT results. As a second test for this problem, a Nelder and Mead [Refs 5.11, 5.12] (NM) multi- dimensional point improvement simplex method (Amoeba) was included in ROKOPT. It was necessary to implement penalty functions to be able to specify constraints in NM as they are not integral to this method. When using NM, it was found that, similar to the EA method, the solutions found depended on the random starting point chosen. This should not be the case for NM and seems to indicate that the introduction of penalty functions, which come into play at the constraint boundary, is responsible for the algorithm not achieving the minimum of the objective function.

71

Chapter 5 Stack Optimisation

5.3.4 Improvements to the Evolutionary Method While Figure 5.3 shows the values of single variables plotted against the GLOW, it appears as if many of the solutions have not converged to any given value but remain suspended in the variable space. It should be pointed out, however, that this figure is the 2-D projections of a 4-dimensional function (three variables and one objective) and, when the solutions are plotted on a 3-dimensional representation, it is possible to see a more enlightened picture of what is occurring. In Figure 5.4, the result of a ROKOPT run of 3240 solution sets is plotted with the GLOW on the vertical axis and the second- and third-stage masses on the horizontal axes. It is now possible to see that all the solutions fall on a surface. The surface in question is the ∆v=11945.5 m/sec constraint surface. The printed page shows some of the detail of the surface; for example, the valley in the GLOW v third-stage mass plane and the extension along the valley floor in the GLOW v second-stage mass plane. This representation is much more convincing, however, when the graphing software is used to rotate the solution sets in real time so that the shape of the surface can be fully visualised. Figure 5.5 shows the same 3240 evolved solution sets plotted on the stage mass axes (m1, m2, m3). Again, the points fall on the ∆v constraint which is more clearly seen on an active display where rotation reveals the nature of the surface. The GLOWs are not shown but can be visualised as contours on the ∆v = constant surface with the minimum GLOW as a point in the centre of the contours at the position (m1, m2, m3) = (220764, 86455, 42589) = 356006kg. Appendix A1 & A2 show the ROKOPT input and output files for this run. Each point on the graph is a single run of ROKOPT, starting with a population of 50 random solutions, and each run has different evolution parameters. What is now apparent is not only that the converged solutions are spread out in variable space but also that they have selectively chosen to sit very accurately on the constraint surface rather than find their way down into the bottom of the valley to the absolute minimum of the objective function. Examination of the evolution history of the populations shows that, as soon as one member of a population arrives at the constraint surface, there is a trend among the feasible individuals to move towards feasibility rather than towards an improvement in the objective function. An analogy can be made to a swarm of bees descending into a valley looking for nectar. As soon as one bee finds a plant with nectar on the valley floor, the remaining bees in the swarm home in on that bee and settle on the same plant, regardless of the fact that there may be sweeter nectar lower down in the valley. The fact that the converged solutions sit on the constraint surface is good as far as satisfying the constraint is concerned and, also, for minimising the GLOW, but only until the ∆v constraint is met, as mentioned in section 5.3.2 above. However, it has to be asked why they are not more closely clustered at the absolute minimum of the valley. When the EA method is started, the generated solutions range over a large volume of the N-dimensional variable space. Within the EA at every generation, all feasible solutions are more highly ranked compared to infeasible solutions. This process can be fundamentally questioned as it may be easier and faster for a marginally infeasible solution to reach the global optima compared to a bad feasible individual because, in reality, optimal solutions are known to lie on constraint boundaries.

5.4 DISCUSSION While the exact parameters of the AR44L launch vehicle are probably known only to the mission analysis staff at Arianespace, by using the data from the user’s manual, a stack

72

Chapter 5 Stack Optimisation

Figure 5.4: Evolved GLOW solutions of a run of ROKOPT using 3240 solution sets plotted in a 3-D representation. The GLOW is plotted on the vertical axis while the values of the second- and third-stage masses are plotted on the horizontal axis. The coloured levels have no particular values. They just serve to help illustrate the 3-D shape of the surface on the 2-D page

Figure 5.5: Evolved stage mass solutions of a ROKOPT run using 3240 solution sets. The solutions are plotted as functions of the three stage masses. The evolved points plotted lie on the Delta-V =11945.5 m/sec surface. The GLOWs are shown as coloured contours of solutions on the Delta-V surface. Once again the colours only serve to illustrate the shape of the surface The minimum GLOW is at point (m1, m2, m3) = (220764, 86455, 42589) = 356006kg. The stage masses are slightly different to those obtained in previous run although the minimum GLOW is identical.

73

Chapter 5 Stack Optimisation model of the vehicle has been developed that gives a mass breakdown and enables an ideal velocity (∆v) figure to be calculated. While this value is probably not the true value, it is close enough to allow a stack model of an equivalent straight three-stage vehicle (the HLV) to be developed. In order to minimise the lift-off weight of the HLV, the choice of stage masses for providing the required ∆v capability must be made correctly. To this end, the STAGEX spreadsheet can be used to produce a nearly optimum stage-mass allocation. The ROKOPT program can be used to further optimise the mass allocation between stages and can do it automatically rather than relying on operator intervention. The development of software to model expendable launch vehicles requires multi- disciplinary tasks amongst which trajectory, guidance, propulsion and structure, for example, are as important as the initial vehicle sizing. In fact, all the disciplines are more or less strongly dependent on each other. In order to optimise the vehicle, a multi- disciplinary model must be constructed to optimise the variables of all the disciplines together. As optimising many variables is beyond the ability of a human operator, a reliable, automatic method must be found that not only works efficiently but can also distinguish between different local minima and find the lowest value. When a problem has a large number of constraints, the solution space may be dissected into a number of discrete regions of “disjoint feasible solutions”. In these cases, the use of gradient methods would require the search to be started in each disjoint region. This, in itself, would require knowledge of the location of each region, a property which may change greatly if more or fewer constraints and their values are included. The determination of the regions’ locations would, by itself, entail a partial solution of the whole problem. EAs overcome this problem by filling the entire N dimensional variable space, within the region of interest, with trial solutions from which the best can be chosen. However, a drawback is that, if disjoint regions exist, more trial solutions are required than for a single region as each sub-region must contain sufficient trial solutions to enable evolution to occur. To reach the absolute minimum may then take a very large number of trials but, once the basin of the global minimum within the search region has been found, it will be more efficient to switch to a gradient or grid search method to reach the absolute minimum of the basin. While constructing a stack model is not a particularly complicated task, it is used as a test bed for evaluating optimisation methods before proceeding to more detailed launch vehicle descriptions. Evolutionary optimisation methods are one of the keys to the efficient discovery of the minimum lift-off weight and, hence, the lowest cost. An alternative method is to partition the problem into a number of smaller problems, such as the stack and trajectory that can be optimised independently. As the various disciplines are dependent on each other, the approximate solutions to each discipline should finally be brought together and a full optimisation carried out on the entire vehicle.

74 Chapter 6 Alternative Launch Vehicle Concepts

6.1 LAUNCH VEHICLE VELOCITY BUDGET While an aircraft is essentially a payload translator designed to move cargo from point A to point B at a relatively low speed, a launch vehicle is a payload accelerator. Once the payload has achieved the required velocity above the atmosphere, the launch vehicle’s job is finished and the payload coasts for the remainder of its mission1.

6.1.1 Nominal Velocity Budget Breakdown In order to size a launch vehicle for a given payload, it is necessary to know the nominal velocity budget required for the given mission. The launch vehicle velocity budget has a number of components which can be broken down as follows: • Orbital velocity; • Gravity loss; • Drag loss; • Steering loss; and • Earth rotation assist. Orbital velocity is the velocity at orbit insertion. For a 200km GTO insertion at the perigee, with the apogee at geostationary height, the orbital velocity is equal to 10,239m/s. Gravity loss is the velocity lost during the lifting of the vehicle to orbital height against gravity. It is less for faster-accelerating vehicles at shallower flight-path angles and greater for slower-accelerating vehicles rising steeply or vertically. For example, a launch vehicle the thrust of which is exactly equal to its mass at all times will hover above the launch pad and the total velocity budget will be wasted in gravity loss. For this reason, a space launch vehicle’s initial T/W ratio at first-stage ignition is generally greater than 1:2 in order to minimise gravity loss and less than 1:5 to minimise its structural requirements, especially those due to axial accelerations incurred at the first-stage burnout. Reference 6.1 gives gravity losses for example vehicles between approximately 1100m/s and 1600m/s. Drag loss usually opposes the thrust but only when the angle of attack is zero. Drag is proportional to the cross-sectional area, that is, it is proportional to L2 where L is the characteristic dimension of the vehicle. The mass of the vehicle varies as L3, hence, the drag (de)acceleration (F=Ma @ a=F/M) and the drag velocity loss vary as 1/L. As examples, reference 6.1 gives the drag loss for the Ariane 44L as 135m/s and for the Saturn 5 as only 40m/s. Steering loss arises from the fact that, in order to control the attitude of the launch vehicle during powered flight, the thrust vector must be slightly misaligned to the velocity vector at times. Thus, there is a wasted transverse component of thrust that does not accelerate the vehicle. Reference 6.1 gives steering losses for the Ariane 44L as 38m/s and for the space shuttle as up to 358m/sec. Earth rotation assist arises from the inertial velocity of the launch site and has been discussed in section 3.3.1. It depends on the launch site latitude and the launch azimuth, and can be up to about 465m/s.

1 Ignoring small velocity trims or changes of orbit

75

Chapter 6 Alternative Vehicle Concepts

6.1.2 Reserve Propellant The sum of all the mission velocity components constitutes the nominal mission velocity requirement. Also, the launch vehicle must be able to supply additional velocity to allow for steering losses incurred by counteracting random variable winds during launch and for the off-nominal performance of the launch vehicle itself. Arianespace normally specifies a probability of 99% for reaching orbit with all off-nominal performance parameters at the 3 sigma level although users are allowed to choose flights with a decreased mission command shutdown probability as low as 50%. Reserve propellant must be provided to create the off-nominal velocity margin as needed. In the case of a launch vehicle performing nominally, all the reserve propellant will be unused. The Ariane 44L utilises a burn-to-depletion strategy in the strap-on boosters and the first stage with no reserve propellant. The second and third stages both carry reserve propellant allowing a command shutdown at a pre-determined velocity. At first glance, it may be thought that it is only necessary to carry reserve propellant in the third stage so that the total velocity can be corrected before the third-stage shutdown. However, carrying reserve propellant in the second stage allows the off-nominal performance of the first and second stages to be corrected so that any planned inter-stage coast between the second and third stages can start with the correct velocity. Otherwise, any velocity errors at the start of a long inter-stage coast would result in a large position error at the third-stage ignition. Carrying a second-stage reserve propellant with a velocity command shutdown also means that less reserve propellant is required in the third stage because part of the velocity deficit has already been corrected by the second stage. The extra second-stage propellant is therefore not a one-to-one loss in terms of total vehicle mass; in fact, for a stage-optimised vehicle, a small velocity exchange between two stages should result in no increase in the vehicle’s lift-off weight.

As a qualitative exercise, the reader could consider the situation in which a given vehicle over-performs in the first and second stages but under-performs in the third while a command shutdown is expected to occur at the nominal second-stage velocity. What does this imply about the amount of reserve propellant to be carried in the third stage? How do the statistics of the two separate systems combined compare with those of a single, combined system?

6.2 LAUNCH VEHICLE IDEAL VELOCITY

6.2.1 The Rocket Equation The total velocity budget, made up of the nominal budget, the random wind-steering loss and the vehicle performance reserve, is represented by the total propellant load of the vehicle. In order to calculate the propellant requirement, the vehicle’s velocity gain in each stage must be calculated and summed over all the stages to obtain the total velocity capability of the vehicle.

76

Chapter 6 Alternative Vehicle Concepts

For a vehicle of mass Mign at engine ignition and mass Ma at engine shutdown, with an engine performance figure of Isp, the velocity gain, ∆v, through the burnout is given by the Rocket Equation, viz:

∆v = Ve.ln(R) ...... Eqn 6.1 where Ve is the exhaust velocity and is given by

Ve = g.Isp ...... Eqn 6.2 and R is called the mass ratio and is given by

R = Mign/Ma ...... Eqn 6.3 This equation was originally derived by in 1903 and refers to a rocket in a vacuum with no drag which is distant from the effects of any gravitational forces.

6.2.2 Calculation of the Ariane 44L Ideal Velocity In order to calculate the ∆v of the AR44L vehicle, a mass breakdown and performance figures for all stages of the vehicle must be known. This data can be found in references 6.2, 6.3 and 6.4. All the data sets available differ from each other with no explanation of the reason/s for the values given. On the production line, each individual stage produced must be weighed to obtain its precise weight which is different from its predecessor. The data used here have been obtained from reference 6.4 and should be considered only as those of a representative vehicle. Table 6.1 gives stage mass breakdowns, specific impulses and burn times. Table 6.2 provides a summary of the vehicle mass breakdown. Table 6.3 is a summary of the velocity changes through each of the mass jettison stages of powered flight. Table 6.1: AR44L properties [Refs 6.2. 6.3, 6.4] AR44L Stage Properties Strap-on First Second Third Designation L40/PAL L220 L33 H10-3 Gross mass of each, kg 44000 246000 38900 12950 Inert mass of each, kg 4500 18000 3500 1250 Propellant mass of each, kg 39500 228000 35400 11700 Isp, sec 278.0 278.4 293.5 445.1 Nominal burn time, sec 140 205 126 745 Jettison time, sec 142

Table 6.2: Ariane 44L vehicle mass breakdown [Refs 6.2, 6.3, 6.4]

Item Mass, kg 4 x liquid strap-on boosters 176000 First stage 246000 Second Stage 38900 Third stage 12950 VEB 530 Payload plus adapter(s) 4768 Fairing 900 GLOW 480048

77

Chapter 6 Alternative Vehicle Concepts

Table 6.3: Summary of ideal velocity changes of Ariane 44L launch vehicle Velocity change Stage m/sec Lift-off to strap-on jettison 2928.2 Strap-on jettison to end first-stage burn 1782.9 Second-stage burn to fairing jettison 1046.9 Fairing jettison to end second-stage burn 1713.9 Third-stage burn 4473.6 Total ideal velocity change 11945.5

Table 6.4 on the following page is a tabular breakdown of the vehicle and component masses through powered flight allowing for calculations of the values in Table 6.3 above.

78

Chapter 6 Alternative Vehicle Concepts

dV m/s 2928.2 1782.9 1046.9 1713.9 4473.6 11945.5 11945.5

sec 278.4 278.4 293.5 278.2 278.2 293.5 445.1 Ivac

1.92 1.92 1.81 2.92 2.92 1.44 2.79 Mass Ratio 6548 76048 76048 58048 58048 40348 39448 21748 18248 480048 480048 164116 146116 Vehicle 0 0 0 0 0 0 0 0 0 Prop 158000 158000 0 0 0 0 0 0 0 0 Strap-ons 18000 18000 18000 Struct Struct 0 0 0 0 0 0 0 70068 70068 70068 Prop Prop 228000 0 0 0 0 0 0 First-stage First-stage 18000 18000 18000 18000 18000 down with Delta-V calculation down Struct 0 0 0 0 900 900 900 900 900 900 900 Fairing 0 0 0 Prop 35400 35400 35400 35400 35400 35400 17700 17700 Component Mass, kg kg Mass, Component 0 0 3500 3500 3500 3500 3500 3500 3500 3500 Second-stage Second-stage Struct 0 Ariane 44L stage mass break Ariane 44L stage Prop Prop 11700 11700 11700 11700 11700 11700 11700 11700 11700 11700 Table 6.4: 1250 1250 1250 1250 1250 1250 1250 1250 1250 1250 Third--Stage Third--Stage Struct Struct 530 530 530 530 530 530 530 530 530 530 530 VEB 4768 4768 4768 4768 4768 4768 4768 4768 4768 4768 P/L & adapter adapter Stage Stage Lift-off Lift-off strap- Before jettison on strap-on After jettison First-stage burnout Second-stage ignition fairing Before jettison fairing After jettison Second-stage Shutdown Third-stage ignition Third-stage shutdown Ariane 44L total ideal velocity velocity total ideal 44L Ariane

79

Chapter 6 Alternative Vehicle Concepts

6.3 ALTERNATIVE VEHICLE CONCEPTS In the following alternative vehicle concept trades, the ∆v of each vehicle is taken as that of the representative AR44L design in order to compare each vehicle with the others, including their sizes. Payload, fairing, equipment and adapter weights are all assumed to be equal to those of the AR44L vehicle.

6.3.1 Modelling Candidate Launch Vehicles In order to demonstrate the savings in the lift-off mass by making alternative propellant choices, a number of candidate launch vehicles were modelled and the stages optimised using the ROKOPT program. The first vehicle, the HLV, is the one discussed in chapter 3. It is the direct conversion of the Ariane technology to a straight, three-stage vehicle and utilises the same propellants, with the same structure factors and stage specific impulses as the AR44L vehicle. This is used as a baseline due solely to the desire to move away from hydrazine-based fuels. Consequently, four other vehicles, which utilise combinations of two of the other propellant choices, viz., LOX/LH2 and LOX/HC, are proposed. The hydrocarbon fuels are classified together under the designation HC as there is very little performance difference between them when their densities are taken into account, as discussed in chapter 3. The main hydrocarbon fuels considered are RP1, liquid methane and liquid propane. The remaining parameters to be selected for the four ALV vehicles are the stage specific impulses and the structure factors. These were chosen from historical data on existing launch vehicles and are set out in Table 6.5.

Table 6.5: Parameter selection used for candidate vehicles Vehicle and Propellant Stage HLV ALV-1 ALV-2 ALV-3 ALV-4

LOX/LH2 LOX/HC LOX/LH2 LOX/LH2 LOX/LH2 445.1s 349s 445.1s 445.1s 445.1s 3 0.0965 0.0666 0.09654 0.0965 0.0965 AR4 H10-3 RLA-2 AR4 H10-3 AR4 H10-3 AR4 H10-3 Ref 6.3 Ref 6.8 Ref 6.3 Ref 6.3 Ref 6.3 N2O4/UH25 LOX/HC LOX/HC LOX/LH2 LOX/LH2 293.5s 349s 349s 430s 430s 2 0.0900 0.0666 0.0666 0.0744 0.0744 AR4 L33 RLA-2 RLA-21 AR5 H155 AR5 H155 Ref 6.3 Ref 6.8 Ref 6.8 Ref 6.7 Ref 6.7 N2O4/UH25 LOX/HC LOX/HC LOX/HC LOX/LH2 278.4s 304s 304s 304s 430s 1 0.0732 0.0591 0.0591 0.0591 0.0744 AR4 L220 Saturn 1C Saturn 1C Saturn 1C AR5 H155 Ref 6.3 Ref 6.6 Ref 6.6 Ref 6.6 Ref 6.7 AR4: Ariane4, AR5: Ariane 5, Saturn1C: 1st stage Saturn 5 RLA-2: 2nd stage of Russian RLA heavy lift launch vehicle using RLA-300 engine

The top row in a cell is the propellant, the second the Isp used, the third the structure factor and the fourth the stage used as the representative stage. Although a stage performance using its specific impulse value is relatively easily estimated, its structure factor is less so as there are a number of historical vehicles and stages from which to choose its value.

80

Chapter 6 Alternative Vehicle Concepts

The structure factors among vehicles using the same propellants will not, in reality, be the same because, as the overall sizes of the vehicles will differ, so will the stage sizes, thereby making for differing structure factors. This can be seen in the first and second stages of the AR44L which both use the same propellant but have structure factors of 0.0732 and 0.0900 respectively. Table 6.6 sets out examples of historical launch vehicles’ first stages utilising LOX/Kerosene for comparison.

Table 6.6: Historical LOX/Kerosene first-stage properties (adapted from M.Wade2)

Titan 1-1: Gross Mass: 76,203 kg (167,998 lb). Empty Mass: 4,000 kg (8,800 lb). Motor: 2 x LR87-3. Thrust (vac): 1,467.908 kN (329,999 lbf). Isp(vac): 290 sec. Burn time: 138 sec. Length: 16.00 m (52.00 ft). Diameter: 3.05 m (10.00 ft). Structure factor: 0.0525 Europa-1: Gross Mass: 89,406 kg (197,106 lb). Empty Mass: 6,997 kg (15,425 lb). Motor: 2 x RZ.2. Thrust (vac): 1,672.671 kN (376,031 lbf). Isp(vac): 282 sec. Burn time: 156 sec. Length: 18.75 m (61.51 ft). Diameter: 3.05 m (10.00 ft), Structure factor: 0.07826 Thor DM-19: Gross Mass: 49,340 kg (108,770 lb). Empty Mass: 3,125 kg (6,889 lb). Motor: 1 x LR79-7. Thrust (vac): 758.711 kN (170,565 lbf). Isp(vac): 282 sec. Burn time: 165 sec. Length: 18.42 m (60.43 ft). Diameter: 2.44 m (8.00 ft), Structure factor: 0.06334 Thor DM-21: Gross Mass: 48,354 kg (106,602 lb). Empty Mass: 2,948 kg (6,499 lb). Motor: 1 x MB-3-1. Thrust (vac): 760.643 kN (170,999 lbf). Isp(vac): 285 sec. Burn time: 164 sec. Length: 18.41 m (60.40 ft). Diameter: 2.44 m (8.00 ft), Structure factor: 0.06097 Delta Thor TA: Gross Mass: 49,442 kg (109,000 lb). Empty Mass: 3,175 kg (6,999 lb). Motor: 1 x MB-3-3. Thrust (vac): 866.710 kN (194,844 lbf). Isp(vac): 290 sec. Burn time: 150 sec. Length: 18.41 m (60.40 ft). Diameter: 2.44 m (8.00 ft), Structure factor: 0.06422 Saturn 1C: Gross Mass: 2,286,217 kg (5,040,245 lb). Empty Mass: 135,218 kg (298,104 lb). Motor: 5 x F-1. Thrust (vac): 38,703.160 kN (8,700,816 lbf). Isp(vac): 304 sec. Burn time: 161 sec, Length: 42.06 m (137.99 ft). Diameter: 10.06 m (33.00 ft), Structure factor: 0.05914

Therefore, the results are only indicative and a full structural design must be mapped before a more accurate structure factor can be obtained. Also, a number of other factors will change, thereby adding more uncertainty to the size values obtained. In particular, as the vehicle size changes, the value of the drag loss will change, albeit by a small fraction of the total velocity requirement, thus modifying the correct value of the ∆v to be used. For all stages, the value of the specific impulse used to calculate the ∆v is the vacuum specific impulse, Ivac, as required by the rocket equation. In operation, the effective specific impulse of the first stage is neither Ivac nor the sea-level specific impulse, Isp(SL). As this value depends mainly on the acceleration of the stage, a variable gravity loss between vehicles is introduced into the considerations. It is assumed that the first stages all undergo the same accelerations and the first-stage engine expansion ratios are such that the effective specific impulses of the stages using the same propellant are identical. The various losses in stage performances are taken up in the velocity allowance made for the ∆v required, as discussed at the start of this chapter.

2 M. Wade, various pages from http://www.astronautix.com. Several specific sources from this database are listed in the references section.

81

Chapter 6 Alternative Vehicle Concepts

Therefore, with these qualifications in mind, the final lift-off weight values and mass distributions obtained between the stages, as set out in Table 6.7, will only be used to help select the propellant combinations for the stages of the candidate vehicles.

Table 6.7:.Optimised stage masses and lift-off weights for candidate launch vehicles

Stage Vehicle Stage Masses, kg HLV ALV-1 ALV-2 ALV-3 ALV-4 3 42625 22675 28359 23046 11540 2 86098 106508 100439 142078 32823 1 221086 215728 109702 0 92895 GLOW 356006 351109 244698 171322 143456

6.3.2 Discussion of Candidate Launch Vehicles In Table 6.6, it can be seen that the HLV’s optimised lift-off mass of 356 tonnes is considerably less than that of the AR44L vehicle’s 480 tonnes. The reason for this is historical, in that the AR4 series of launchers was derived from the technology of the AR1 launch vehicle. When Arianespace desired to increase the lifting capacity of the Ariane without developing new engines or tankage components, a non-optimal design, in terms of lift-off weight, was accepted for the subsequent AR2, AR3 and AR4 series of launchers. When a new launch vehicle is to be designed, the conventional wisdom is that the minimum lift-off weight leads to the minimum cost vehicle. However, with current technology, the modification of an existing vehicle demands minimum development cost with minimum changes of tooling and, especially, little engine development. The AR4 was the end of the development line. The Ariane 4 core vehicle, Ariane 40, cannot lift off with a full payload. Due to the need for increased lifting capacity, new engines had to be developed so a totally new launch vehicle design was justified. This resulted in the Ariane 5 launcher. For the vehicles under examination here, which are to be clean-sheet designs, the minimum lift-off weight is a primary design criterion. The ALV vehicles in Table 6.7 show decreasing lift-off weights as the performance of each stage progressively increases. The ALV-3 vehicle is interesting in that the two upper stages are capable of providing the required ∆v on their own without the provision of a third stage. The benefit of a third stage is outweighed by the lower performance of the LOX/HC propellant combination compared to the LOX/LH2 propellant in the upper stages. Hence, the optimum weight for the first stage is zero. In the case of the ALV-4 vehicle, the GLOW is further reduced as the three-stage concept is again of benefit because all stages utilise the LOX/LH2 propellant.

82

Chapter 6 Alternative Vehicle Concepts

6.4 CONSTRAINTS ON STACK OPTIMISATION Although the stage optimisations above give minimum GLOW values for the design parameters selected, there are a number of constraints on the designs which must be examined. They are: • Maximum Load Factor; • Stage Physical Size; and • Engine Commonality.

6.4.1 Load Factor Considerations The maximum load factor usually occurs at the first-stage burnout. This is because the thrust of the first-stage engines must be chosen to be at least equal to the lift-off weight. As the first-stage lift-off is influenced by any wind present, the higher the lift-off thrust, the higher the acceleration and the less the effect of the wind on the vehicle. Furthermore, the greater the acceleration, the lower is the gravity loss. For this reason, the first-stage T/W ratio is usually chosen to be greater than 1.2 [Ref 6.9]. While an even higher acceleration is desirable, it is limited to lower than about 1.5 for structural (load factor) reasons relevant to both the vehicle airframe and the payload. As the altitude of the vehicle increases, the ambient air pressure decreases thereby causing the thrust of the engines to increase. As the vehicle mass decreases, the acceleration and load factor increase until the first-stage burnout. Figure 6.1, adapted from the Ariane 4 User’s Manual, illustrates the flight load factor history of the AR44LP vehicle. The maximum load factor is limited to 4.0g for the benefit of the payload.

Figure 6.1: Ariane 44LP trajectory – load factor versus time (Adapted from Ariane-4 User’s Manual)

83

Chapter 6 Alternative Vehicle Concepts

For the calculations shown in Table 6.7, ROKOPT optimised away the first stage of the ALV-3 vehicle. The calculations were re-done, replacing stages 2 & 3 as 1 & 2. The stage-optimised HLV’s and ALV’s values are shown in Table 6.8 with an initial T/W ratio of 1:3 and an allowance of 10% for the increase of thrust from sea level to vacuum. The load factors at the first-stage burnout were calculated and all vehicles demonstrated a maximum load factor below the 3.8g of the ALV-4, except for the ALV-3 which had a maximum load factor of 6.7g. This high value is typical of two-stage vehicles for which the higher fractional propellant burnoff per stage increases the acceleration per stage. To improve the load factor situation of the ALV-3, an off-optimal stage mass distribution could be adopted. Table 6.9 shows the ALV-3 vehicle with a number of second-stage masses chosen. The first stage is then matched to give the required 1145.5m/s ∆v and the procedure shown in Table 6.8 is repeated to obtain the load factors. A first-stage five- engine cluster is assumed, with all the engines operating at lift-off and the centre engine able to be shut down late in the first-stage burnout to reduce the maximum load factor where required. The last two rows in Table 6.9 show the maximum load factors of five and four engines operating at stage burnout respectively.

Table 6.8: Calculation of first-stage burnout load factor

Parameters HLV ALV-1 ALV-2 ALV-3 ALV-4 GLOW, kg 356006 351109 244698 194137 143456 1st stage mass, kg 221086 215728 109702 162467 92895 1st stage structure factor 0.0732 0.0591 0.0591 0.0744 0.0744 1st stage propellant mass, kg 204903 202978 103219 150379 85984 Vehicle 1st stage BO mass, kg 151103 148131 141479 43758 57472 Initial thrust to weight 1.3 1.3 1.3 1.3 1.3 1st stage SL thrust, kgf 462808 456442 318107 252378 186493 ~1st stage vac thrust, kgf 518873 524908 365824 290235 214467 1st stage max load factor, g 3.43 3.54 2.59 6.63 3.73

The calculations show that a second-stage mass of 44.0 tonnes (44Mg) and a first-stage mass of 156.7 tonnes gives a GLOW of 206.9 tonnes, a penalty of 12.8 tonnes (6.6%) but a maximum load factor of 5.0g with five engines operating, and 4.0g with four engines operating, at the first-stage burnout. As the modified ALV-3’s GLOW still fell between that of the ALV-2 and ALV-4 vehicles, load factor considerations did not seriously affect the two- stage ALV-3 concept.

84

Chapter 6 Alternative Vehicle Concepts

Table 6.9: Calculation of off-optimal ALV-3 first-stage burnout load factor Parameters Off optimal ALV-3 versions 2nd stage mass, Mg 25.473 30.000 40.000 44.00 50.00 60.00 70.00 80.00 90.00 1st stage mass, Mg 162.47 159.08 156.59 156.69 157.50 160.06 163.54 167.60 172.03 GLOW, Mg 194.14 195.28 202.79 206.88 213.70 226.26 239.74 253.80 268.23 1st stage structure factor 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 0.074 1st stage propellant, Mg 150.38 147.24 144.94 145.03 145.78 148.15 151.38 155.13 159.23 Vehicle S1 BO mass, Mg 43.76 48.033 57.849 61.85 67.92 54.64 88.37 98.67 109.00 Initial thrust to weight 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1st stage SL thrust, Mgf 252.38 253.86 263.63 268.95 277.81 294.13 311.66 329.93 348.70 ~1st stage vac thrust, Mgf 290.24 291.94 303.17 309.29 309.29 319.48 338.25 358.41 379.42 S1 5 eng max load factor, g 6.63 6.08 5.24 5.00 4.70 4.33 4.06 3.85 3.68 S1 4 eng max load factor, g 5.31 4.86 4.19 4.00 3.76 3.46 3.24 3.08 2.94

6.4.2 Stage Physical Size The handling and transport of the large tanks required for the stages (in particular, the first) are important considerations in the vehicle development program and during the manufacture and operation of the launch system. The physical sizes of the tanks for the first stages of the candidate vehicles were calculated from their propellant weights in Table 6.9 and their propellant loading ratios. Propellant Loading Ratio: The relevant mixture ratios for stage sizing were those of the stage propellant loadings rather than the combustion chamber mixture ratio. The total propellant load was made up of usable and unusable propellants. The unusable propellant is that left in the tanks, pipes and turbopump at engine shutdown. The usable propellant is divided between the combustion chamber supply for the production of thrust and about 10% of the total propellant used to run the turbopump gas generator which runs at a much greater fuel- rich ratio than does the main combustion chamber. Therefore, the stage propellant loading ratio was different to the mixture ratio at which the combustion chamber ran. Examples of the stage propellant loadings’ historical stages are given in Table 6.10. Table 6.10: Stage propellant loadings and ratios [Refs 6.12, 6.13] Stage Oxidizer Fuel Engine Stage Propellant Loading Engine Load, Kg Load, Kg MR Ratio S-IC LOX/RP1 1423982 614257 2.32 F1 2.27

S-II LOX/LH2 360961 70300 5.13 5 x J2 5.50

S-IVB LOX/LH2 86891 19705 4.41 1 x J2 5.50

H10-III LOX/LH2 9117 2022 4.51 HM-7B 6.00 Notes: 1. LOX/RP1 candidate stages use propellant loading ratios of the S-1C stage. 2. LOX/Propane candidate stages use propellant loading ratios of the S-1C stage modified by the ratio of LOX/RP1 to LOX/Propane combustion Max Isp mixture ratio

85

Chapter 6 Alternative Vehicle Concepts

The candidate ALV stages using LOX/RP1 were set to have the loading ratio of the Saturn-1C stage while those using LOX/Propane, although set to the same ratio, were adjusted by the ratio of the combustion mixture ratios for max Isp for LOX/Propane and LOX/RP1 (=2.8/2.6), as determined by NASA CEA calculations (Chapter 4). The candidates’ second stages, using LOX/LH2, were set to have the propellant loading ratio of the AR4 H10-III stage. Tank Shape Aspect Ratios (L/D): For calculating the tank sizes, the tanks were treated as right-circular cylinders whereas, in reality, they are produced with elliptical ends. Thus, the sizes are only indicative for making comparisons among vehicles. The aspect ratios of the tanks were chosen from those of historical launch vehicles. While the HLV vehicle has the same propellants as the AR4, its L/D was not chosen to be the same as that of the AR4 because the L220 stage tank ratios had high aspect ratios, due to the maximum possible tank diameter in the AR4 system. The upgrade from AR1 to AR4 relied upon lengthening the tanks with no corresponding increases in their diameters. The HLV first-stage tank L/D was, therefore, set to be the same as the LOX/RP1 Saturn 5 first stage (S-1C). The tank layouts of the three Saturn 5 stages are shown in Figures 6.2, 6.3 & 6.4 and their L/D ratios are tabulated in Table 6.11.

Table 6.11: Saturn 5 stage tank aspect ratios

Saturn-V Diameter Length Tank Propellant Stage (inches) (inches) L/D S-IC LOX/RP-1 396 1286 3.25

S-II LOX/LH2 396 936 2.36

S-IVB LOX/LH2 260 528 2.00

86

Chapter 6 Alternative Vehicle Concepts

Figure 6.2: Saturn-V S-1C first stage structure with tank dimensions [Reproduced from reference 6.12]

87

Chapter 6 Alternative Vehicle Concepts

Figure 6.3: Saturn-V S-II second stage structure with tank dimensions [Reproduced from reference 6.12]

88

Chapter 6 Alternative Vehicle Concepts

Figure 6.4: Saturn-V S-IVB third stage structure with tank dimensions [Reproduced from reference 6.12]

89

Chapter 6 Alternative Vehicle Concepts

Other candidate ALV stages using LOX/RP1 were also set to have the same L/D as S-1C while the L/D of the LOX/Propane first stages were set as equal to those of the Saturn-V S- II’s second stage. Candidate ALV LOX/LH2 stages were set with L/Ds equal to the Saturn-V SIVB’s third stage. Differences were due to the insulation required because of temperature differences between the propellants, as discussed in Chapter 4. Table 6.12 sets out the calculated sizes of the first-stage tanks for the AR4, HLV and ALV-1 to ALV-4 vehicles. The effects on the tank size of adopting two other tank diameters, in addition to the exemplar stage L/D, are tabulated in Table 6.12 using diameters of 4.0m and 5.0m. For comparison, the Ariane 1 to 4 first-stage diameters were 3.8m while the Ariane 5 LOX/LH2 common-core stage had a diameter of 5.4m.

Table 6.12: Candidate vehicles’ first-stage tanks’ physical sizes AR4 HLV ALV-1 ALV-2 ALV-3 ALV-4 L220

LOX LOX LOX LOX LOX LOX N2O4/UH25 Propane RP-1 Propane RP-1 LH2 LH2 Density Fuel, kg/m3 829 829 790 810 790 810 70 70 Density Oxidiser, kg/m3 1477 1477 1140 1140 1140 1140 1140 1140 O/F Loading Ratio 1.70 1.70 2.50 2.32 2.50 2.32 4.51 4.51 Density Propellant, kg/m3 1145 1145 1012 1015 1012 1015 302 302 Stage Mass, kg 246000 221086 215728 215728 109702 109702 156685 92895 Structure Factor 0.0732 0.0732 0.0591 0.0591 0.0591 0.0591 0.0591 0.0744 Mass Structure, kg 18000 16183 12750 12750 6483 6483 9260 6911 Mass Propellant, kg 228000 204903 202978 202978 103219 103219 147425 85984 Tank Volume, m3 199.1 178.9 200.6 199.9 102.0 101.7 488.1 284.7 Type 1: Tank diameter based on exemplar stage L/D Tank Structure L/D Ratio 4.62 3.25 2.36 3.25 2.36 3.25 2.00 2.00 Tank Diameter, m 3.80 4.12 4.77 4.28 3.80 3.41 6.77 5.66 Tank X-Area, m2 11.34 13.35 17.84 14.38 11.36 9.16 36.03 25.15 Tank Length, m 17.55 13.40 11.25 13.90 8.98 11.10 13.55 11.32 Type 2: Tank diameter = 4.0m Tank Structure L/D Ratio 3.55 3.56 3.99 3.98 2.03 2.02 9.71 5.66 Tank Diameter, m 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 Tank X-Area, m2 12.57 12.57 12.57 12.57 12.57 12.57 12.57 12.57 Tank Length, m 15.81 14.24 15.96 15.91 8.12 8.09 38.84 22.65 Type 3: Tank diameter = 5.0m Tank Structure L/D Ratio 2.03 1.82 2.04 2.04 1.04 1.04 4.97 2.90 Tank Diameter, m 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 Tank X-Area, m2 19.63 19.63 19.63 19.63 19.63 19.63 19.63 19.63 Tank Length, m 10.14 9.11 10.22 10.18 5.20 5.18 24.86 14.50 Notes: 1. All tanks treated as right-circular cylinders for comparisons only 2. AR4 L220 1st-stage tank length from ref 6.11 given as 18.4m (includes elliptical ends) 3. ALV-3 is the off-optimal version with S2 mass = 44 tonnes 4. Propellant densities (except propane at LOX temps) from ref 6.10

90

Chapter 6 Alternative Vehicle Concepts

The diameters and lengths of the stages are important not only for manufacturing and transport but for matching them to the adjacent stages, S1 to S2 and S2 to S3 respectively. The payload fairing diameter should also be sufficient to accept prospective payloads while being approximately matched to the maximum diameter of the stages. Figure 6.5 shows diagrams of the Ariane 4 payload compartments for a dual spacecraft launch with the fairing envelope indicated. The longest fairing available was 12.4m in length. The physical sizes of the second and third stages were also calculated and are tabulated in Tables 6.13 and 6.14.

Figure 6.5: Ariane 4 payload compartment configurations for dual spacecraft launches. Maximum fairing length 12.4m for long fairing [Reproduced from reference 6.2]

If the stage aspect ratio became too low, the command authority of the thrust vectoring of the engines would become a problem on the shorter arms between the C of G and the engine. While the high performance of the LOX/Hydrogen stages decreased the lift-off weight of the vehicle, the very low density of the liquid hydrogen made the average fuel density very low so that the tank size was large. Thus, LOX/Hydrogen is of no advantage in reducing the vehicle size.

91

Chapter 6 Alternative Vehicle Concepts

Table 6.13: Candidate vehicles’ second-stage tanks’ physical sizes

AR-L33 HLV ALV-1 ALV-2 ALV-3 ALV-4

LOX LOX LOX LOX N O /UH25 LOX/LH 2 4 Propane RP-1 Propane RP-1 2 Density Fuel, kg/m3 829 829 790 810 790 810 70 70 Density Oxidiser, kg/m3 1477 1477 1140 1140 1140 1140 1140 1140 O/F Loading ratio 1.70 1.70 2.50 2.32 2.50 2.32 4.51 4.51 Density Propellant, kg/m3 1145 1145 1012 1015 1012 1015 302 302 Mass Stage, kg 38900 86098 106508 106508 100439 100439 44000 32823 Structure Factor 0.0900 0.0900 0.0666 0.0666 0.0666 0.0666 0.0744 0.0744 Mass Structure, kg 3500 7749 7093 7093 6689 6689 3274 2442 Mass Propellant, kg 35400 78349 99415 99415 93750 93750 40726 30381 Tank Volume, m3 30.9 68.4 98.2 97.9 92.6 92.3 134.8 100.6 Type 1: Tanks with same diameter as first stage Tank Structure L/D Ratio 0.72 1.24 1.16 1.59 2.14 2.95 0.55 0.71 Tank Diameter, m 3.8 4.12 4.77 4.28 3.80 3.41 6.77 5.66 Tank X-Area, m2 11.34 13.35 17.84 14.38 11.36 9.16 36.03 25.15 Tank Length, m 2.73 5.12 5.51 6.81 8.15 10.08 3.74 4.00 Type 2: Tank diameter equal to 4.0m Tank Structure L/D Ratio --- 1.36 1.95 1.95 1.84 1.84 2.68 2.00 Tank Diameter, m --- 4.00 4.00 4.00 4.00 4.00 4.00 4.00 Tank X-Area, m2 --- 12.57 12.57 12.57 12.57 12.57 12.57 12.57 Tank Length, m --- 5.44 7.82 7.79 7.37 7.35 10.73 8.00 Type 3: Tank diameter equal to 5.0m Tank Structure L/D Ratio --- 0.70 1.00 1.00 0.94 0.94 1.37 1.02 Tank Diameter, m --- 5.00 5.00 5.00 5.00 5.00 5.00 5.00 Tank X-Area, m2 --- 19.63 19.65 19.65 19.65 19.65 19.65 19.65 Tank Length, m --- 3.48 5.00 4.98 4.72 4.70 6.86 5.12 Notes: 1. Propellant densities (except propane at LOX temps) from ref 6.10 2. AR4 second-stage tank sizes are those of the actual vehicle 3. ALV-3 is off-optimal version with S2 mass = 44 tonnes

6.4.3 Choice of Engine Number in First-Stage Cluster When the number of engines in the first-stage cluster is to be determined, certain factors must be considered, in particular, the greater the number of engines to be used, the more the complexity and the slightly higher the weight. However, as more engines provide some redundancy for an engine failure during flight and mean a lower thrust per engine, this leads to physically smaller engines with consequential savings in development costs. A single engine means no redundancy, high thrust and large size and weight. Two engines (e.g., Titan, Blue Streak) are an improvement in this regard but produce an asymmetric thrust vector control in pitch and yaw.

92

Chapter 6 Alternative Vehicle Concepts

Three engines are totally asymmetric in pitch and yaw (an in-line three-engine cluster can be equivalent to a two-engine cluster for control, e.g., Atlas) while a four-engine cluster is totally symmetric. The addition of a non-vectoring centre engine in a five-engine cluster improves redundancy over that of a four-engine cluster and lowers the thrust, weight (per engine) and size requirements. Six- and seven-engine clusters are non-symmetric in pitch and yaw control. Eight-engine clusters can be constructed with four outer vectored engines and four inner fixed engines (e.g., Saturn-1B). From there, an additional fixed engine can be placed in the centre to create a nine-engine cluster. Beyond this figure, conventional engine clusters become too complex and weighty3. For these reasons, for the three-stage ALV candidates, a five-engine cluster (as per Saturn-1C and Saturn-II) was considered to be the most practical.

Table 6.14: Candidate vehicles’ third-stage tanks’ physical sizes AR4- HLV ALV-1 ALV-2 ALV-4 H10+

LOX/ LOX/RP LOX/LH LOX/LH 2 Propane 1 2 Density Fuel, kg/m3 70 70 790 810 70 70 Density Oxidiser, kg/m3 1140 1140 1140 1140 1140 1140 O/F Loading Ratio 4.51 4.51 2.50 2.32 4.51 4.51 Density Propellant, kg/m3 302 302 1012 1015 302 302 Stage Mass, kg 12950 42625 22675 22675 28359 11540 Structure Factor 0.0965 0.0965 0.0666 0.0666 0.0965 0.0965 Mass Structure, kg 1250 4113 1510 1510 2737 1114 Mass Propellant, kg 11700 38512 21165 21165 25622 10426 Tank Volume, m3 38.7 127.5 20.9 20.8 84.8 34.5 Type 1: Tank diameter based on exemplar stage L/D Tank Structure L/D Ratio 2.81 2.00 2.36 2.00 2.00 2.00 Tank Diameter, m 2.60 4.33 2.24 2.37 3.78 2.80 Tank Length, m 7.30 8.66 5.29 4.73 7.56 5.60 Type 2: Tank diameter 3.0m (& 4.0m) Tank Structure L/D Ratio ------2.39 2.40 1.86 1.45 Tank Diameter, m ------3.00 3.00 4.00 3.00 Tank Length, m ------2.96 2.95 6.75 4.88 Notes: 1. Ariane 3rd-stage (H10-3) tank sizes from refs 6.11, 6.13, 6.14

6.4.4 The Stacks, Load Factors & Engine Commonality Table 6.15 sets out the stack properties for each of the reference and candidate vehicles. The masses of the vehicle equipment bay, adapter and fairing make up the 6198kg mass after jettison of the last stage. From these calculated masses, the load factors due to engine thrust can be obtained. This enables engine commonality between the stages, as well as structural limits, to be determined.

3 The ill-fated Russian 5-stage moon launch vehicle had 30 NK-15 engines clustered around its base.

93

Chapter 6 Alternative Vehicle Concepts

Table 6.15: Summary of candidate vehicle masses – the stacks

AR4 HLV ALV-1 ALV-2 ALV-3 ALV-4 THE STACK Vehicle Mass at S1 ignition, kg 480048 356006 351109 244698 206883 143456 S1 Mass, kg 246000 221086 215728 109702 156685 92895 S1 Propellant Mass, kg 228000 204903 202978 103219 147425 85984 Vehicle Mass at S1 burnout, kg 76048 151103 148131 141479 59458 57472 S1 Structure Mass, kg 18000 16183 12750 6483.4 9260 6911.4 Vehicle Mass at S2 ignition, kg 58048 134920 135381 134996 50198 50561 S2 Mass, kg 38900 86098 106508 100439 44000 32823 S2 Propellant Mass, kg 35400 78349 99415 93750 40726 30381 Vehicle Mass at S2 burnout, kg 22648 56571 35966 41246 9472 20180 S2 Structure Mass, kg 3500 7748.8 7093.4 6689.2 3274 2442 Vehicle Mass at S3 ignition, kg 19148 48822 28873 34557 n/a 17738 S3 Mass, kg 12950 42625 22675 28359 n/a 11540 S3 Propellant Mass, kg 11700 38512 21165 25622 n/a 10426 Vehicle Mass at S3 burnout, kg 7448 10310 7708 8935 n/a 7312 S3 Structure Mass, kg 1250 4113 1510 2737 n/a 1114 Vehicle - last stage structure, kg 6198 6198 6198 6198 6198 6198

As is to be expected, the vehicles’ GLOWs decreased as the number of high-performing LOX/LH2 stages increased in the candidate vehicles, with the ALV-4 being only 41% of the mass of the ALV-1. However, as will be seen in the vehicle trade summaries in section 6.6, the sizes of the vehicles did not change dramatically due to decreases in their GLOWs. From the data presented, it can be seen that, for reasonable load factor limits of the first stage, the ALV-3’s second stage suffered from excessive load factors when equipped with either one or two of the first-stage engines. The problem is due to the ratios of the vehicle masses at the first- and second-stage ignitions. Although an off-optimal solution which satisfies the first-stage load factors (with five engines) was adopted, the second-stage load factors were still not satisfied. The second-stage mass needed to be increased further, while the first-stage mass was correspondingly decreased, until the load factors could be satisfied. Thrust levels were assumed to increase by 10% from sea level to vacuum in the first stage and by an additional 3% through the addition of a lengthened skirt on the second-stage engine nozzle. ALV-1, -2 & -4 all had satisfactory load factors (Table 6.16) for their first and second stages. As launch vehicles’ third stages commonly operate at accelerations around, or lower than, 1g, for the three-stage ALVs, the third-stage engine thrust was chosen arbitrarily to give reasonable T/W ratios. A vacuum thrust level of between 6,000 Kgf and 10,000 Kgf was seen to be satisfactory.

94

Chapter 6 Alternative Vehicle Concepts

Table 6.16: Load factors for engine commonality & structure

AR4 HLV ALV-1 ALV-2 ALV-3 ALV-4 FIRST STAGE

S1 Propellant N2O4/UH25 LOX/HC LOX/LH2 S1 5 Eng = SL Thrust per, kgf 69070 92562 91288 63621 53790 37299 S1 T/W 1.15 1.3 1.3 1.3 1.3 1.3 S1 5 Engine Vac Thrust n/a 509089 502086 349918 295843 205142 S1 5 engine T/W @ burnout n/a 3.44 3.39 2.47 4.98 3.57 S1 4 Engine Vac Thrust 309592 407271 401669 279935 236674 164114 S1 4 engine T/W @ burnout 4.07 2.70 2.71 1.98 3.98 2.86 SECOND STAGE

S2 Propellant N2O4/UH25 LOX/HC LOX/LH2 S2 2 Eng Vac Thrust, kgf n/a 210763 207864 144866 122479 84929 S2 2 Eng Init T/W n/a 1.56 1.54 1.07 2.44 1.68 S2 2 EngT/W @ burnout n/a 3.73 5.78 3.51 12.9 4.21 S2 1 Eng Vac Thrust, kgf 80102 105381 103932 72433 61239 42464 S2 1 Eng Init T/W 1.38 0.78 0.77 0.54 1.22 0.84 S2 1 Eng T/W @ burnout 3.54 1.86 2.89 1.76 6.47 2.10 THIRD STAGE

S3 Propellant N2O4/UH25 LOX/HC LOX/LH2 S3 1 Eng Vac Thrust, kgf 6398 10000 10000 10000 n/a 6000 S3 1 Eng Init T/W 0.33 0.20 0.35 0.29 n/a 0.34 S3 1 EngT/W @ Burnout 0.86 0.97 1.30 1.12 n/a 0.82

6.5 OPTIMISATION OF TWO-STAGE VEHICLE FOR STAGE LOAD FACTORS In order to test whether the ALV-3 vehicle could be stage-optimised to allow engine commonality, the stage masses were adjusted while the required total ∆v was maintained. Given the mass of the vehicle above the upper stage and the data for the second-stage structure factor and specific impulse, the second-stage ignition and burnout masses were calculated for a given second-stage mass. The second-stage velocity change was then easily calculated as was the first-stage velocity change. Knowing the first-stage specific impulse, the required first-stage mass ratio could be easily calculated. Then, with M2I = Mass of second stage at ignition,

M1B = Vehicle mass at first-stage at burnout,

M1I = Vehicle mass at first-stage ignition,

MS1 = First-stage structure mass,

MP1 = First-stage propellant mass, R = First-stage structure mass ratio,

Sf = First-stage structure factor,

MS = First-stage mass,

95

Chapter 6 Alternative Vehicle Concepts

and having …

M1B = M2I+MS1,

M1I = M2I+MS1+MP1,

R = M1I/M1B,

R = (M2I+MS1+MP1)/(M2I+MS1),

MS1+MP1 = MS,

MS1 = MS·Sf,

R = (M2I+MS)/(M2I+MS·Sf),

R·(M2I+MS Sf) = (M2I+MS),

R·MS·Sf+R.M2I = M2I+MS,

MS·(1-R·Sf) = M2I·(R-1) and

MS = M2I·(R-1)/(1-R Sf), …the mass of the first stage could be calculated. The vehicle would now deliver the required ∆v and the engine thrust of the first stage could be calculated for a given ignition T/W ratio. Applying this engine to the second stage, with a 12%4 increase in thrust from SL to vacuum and a 3.5%5 increase due to an extended nozzle skirt (expansion ratio), the ignition and burnout T/W ratios for the second stage were calculated. Table 6.17 shows the results of these calculations being carried out with the second-stage mass being an independent variable. The result is that, for second-stage masses ranging from 40,000kg to 160,000kg, the second- stage T/W ratio never fell within the maximum 4.0g range. These second-stage weights corresponded to a vehicle GLOW ranging from 202,936 kg to 374,236 kg - an increase of over 84%. It is concluded that engine commonality is not possible for a two-stage vehicle without a throttleable second-stage engine (that is, a S1 engine), or a S1 ignition T/W ratio of below 1:1 and a maximum load factor higher than 4.0g being accepted6.

4 Value equals increase in Ariane 4 S1 Viking VC engine SL to Vac 5 Value equals increase between Ariane 4 S1 Viking VC engine and S2 Viking IVB in vacuum 6 Ariane 4: S1 T/Wign = 1.16; Ariane 5: lift-off T/W= 1.5, S1 (EPC, H155) burnout load factor = 4.76g

96

Chapter 6 Alternative Vehicle Concepts

Table 6.17: Engine commonality of two-stage launch vehicle

ALV-3 Vehicle Variants

Second Stage Mass, kg 44000 60000 80000 100000 120000 140000 160000 Structure Factor 0.0965 0.0965 0.0965 0.0965 0.0965 0.0965 0.0965 Propellant Mass, kg 39754 54210 72280 90350 108420 126490 144560 Structure Mass, kg 4246 5790 7720 9650 11580 13510 15440 Burnout Mass, kg 10444 11988 13918 15848 17778 19708 21638 Ignition Mass, kg 50198 66198 86198 106198 126198 146198 166198 Mass Ratio 4.81 5.52 6.19 6.70 7.10 7.42 7.68 Specific Impulse, sec 445.1 445.1 445.1 445.1 445.1 445.1 445.1 ∆V, m/sec 6852.3 7458 7958 8302 8554 8746 8898

First Stage S1 ∆V Required, m/s 5093.2 4487 3986 3642 3391 3199 3047 Specific Impulse, sec 430.0 430.0 430.0 430.0 430.0 430.0 430.0 Mass Ratio Required 3.3 2.9 2.6 2.4 2.2 2.1 2.1 Structure Factor 0.0744 0.0744 0.0744 0.0744 0.0744 0.0744 0.0744 Stage Mass, kg 156837 160242 167821 176992 186950 197355 208038 Propellant Mass,kg 145168 148320 155335 163823 173041 182672 192560 Structure Mass, kg 11669 11922 12486 13168 13909 14683 15478 S1 Ignition Mass, kg 207035 226440 254019 283190 313148 343553 374236 S1 Burnout Mass, kg 61867 78120 98684 119366 140107 160881 181676 Mass Ratio 3.3 2.9 2.6 2.4 2.2 2.1 2.1

GLOW, kg 207035 226440 254019 283190 313148 343553 374236 S1 5-Eng Ignition T/W 1.3 1.3 1.3 1.3 1.3 1.3 1.3 SL Thrust, kgf 269145 294372 330224 368146 407092 446619 486507 Vac Thrust Increase 12% 12% 12% 12% 12% 12% 12% 5-Eng Vac Thrust, kgf 301442 329697 369851 412324 455943 500213 544888 S1 5-Eng Burnout T/W 4.9 4.2 3.7 3.5 3.3 3.1 3.0 4-Eng Vac Thrust, kgf 241154 263757 295881 329859 364755 400171 435910 S1 4-Eng Burnout T/W 3.9 3.4 3.0 2.8 2.6 2.5 2.4

S2 Ignition Mass, kg 50198 66198 86198 106198 126198 146198 166198 S2 Burnout Mass, kg 10444 11988 13918 15848 17778 19708 21638 S2 Thrust Increase S1-S2 3.5% 3.5% 3.5% 3.5% 3.5% 3.5% 3.5% S2 Eng Thrust, kgf 62406 68255 76568 85361 94391 103556 112804 S2 1-Eng Ignition T/W 1.2 1.0 0.9 0.8 0.7 0.7 0.7 S2 1-Eng Burnout T/W 6.0 5.7 5.5 5.4 5.3 5.3 5.2 Notes: 1. Engine commonality can be achieved with S1 ignition T/W ratio around 1:1 and a burnout T/W ratio of 5.0 accepted on S2. 2. Ariane 5 T/W ratio at S1 common core burnout (EPC, H155) = 4.76 [Ref 6.3]

97

Chapter 6 Alternative Vehicle Concepts

6.6 CANDIDATE VEHICLE TRADE DISCUSSION A trade table, (Table 6.18) containing the physical properties of the four candidate vehicles, can now be constructed. There is seen to be very little difference between the RP1 and propane stages so all the hydrocarbon stages can be considered together. As the number of LOX/hydrogen stages increased, from ALV-1 to -4, the lift-off weight of the vehicle decreased. With two LOX/Hydrogen stages, a third stage was not necessary for the GTO mission. Adding a third LOX/Hydrogen stage decreased the lift-off weight to below that of the two-stage vehicle and decreased the size of the vehicle although it was still larger than the hydrocarbon-fuelled vehicles. Apart from the two-stage vehicle, the size differences are perhaps not significant. More important considerations are the design, manufacturing and operational aspects of the number of stages, the number of propellant technologies to be mastered and the possible engine commonalities between stages, i.e., the number of engine designs required to be developed. Considerations such as manufacturing simplicity, handling and transport, and overall operational complexity should be considered above other factors. These matters have been outlined in Chapter 3. Table 6.18: Trade table for candidate vehicles

ALV-1 ALV-2 ALV-3 ALV-4 Oxidizer LOX LOX LOX LOX

Fuel LHC LHC, LH2 LH2 LH2 GLOW, kg 351109 244698 206883 143456 No of Stages 3 3 2 3 No Engines S1 5 5 5 5 No Engines S2 1 1 1 1 No Engines S3 1 1 n/a 1 No of Engine Technologies 1 2 1 1 No of Engines to be Developed 2+ 2+ 2 2+ S1/S2 Engine Commonality Possible Yes Yes No? Yes S1 Tank Diameter, m 5.0 5.0 5.0 5.0 S1 Tank Height, m 10.2 5.2 24.9 14.5 S2 Tank Diameter, m 4.0 4.0 4.0 4.0 S2 Tank Height, m 7.8 7.4 10.7 8.0 S2 Tank Diameter, m 3.0 4.0 n/a 3.0 S3 Tank Height, m 3.0 6.8 n/a 4.9 S3 Total Tankage Height, m 21.0 19.4 35.6 27.4 Notes: ALV-3 requires S2 throttleable, S1 one engine to shut down (Table 6.17) A + under engine development means a modification to second-stage engine Ariane 5 common core (EPC, H155) OD = 5.46, S1 stage length = 30.5 m, S2 stage (EPS, L9) length = 3.36m

98

Chapter 6 Alternative Vehicle Concepts

Figure 6.6 shows the layout of the ALV-4 vehicle with its approximate dimensions.

Figure 6.6: Layout of candidate vehicle ALV-4 (Dimensions are approximate to give indication of scale only.)

99

Chapter 6 Alternative Vehicle Concepts

6.7 SENSITIVITY OF STAGE IDEAL VELOCITY TO SPECIFIC IMPULSE & MASS RATIO The above calculations were made with values for the stage structure factors and the engine specific impulses taken from historical data. The specific impulses are relatively easily estimated as they are determined by the physical properties of the propellants however the stage structures are less easily estimated as they depend on the engineering employed. Hence the results of the calculations could be criticized on the basis of the accuracy of the input values being employed. There are two reasons why this is not correct. Firstly, the important factor is to be able to make a comparison between the alternative vehicles rather than to determine an exactly correct lift-off weight. Secondly the stage masses are chosen to give a set vehicle ideal velocity. The achieved velocity is calculated through the rocket equation [Equation 6.1]. The velocity is directly proportional to the specific impulse but only proportional to the log of the mass ratio which is determined by the structure factor, viz:

∆v = Ve.ln(R) the structure factor is

Fs = Ms/(Ms+Mp)

Where Ms is the structure mass and Mp is the propellant mass. But,

R = (Mp+Ms)/Ms Hence the stage mass ratio is the reciprocal of the structure factor

R = 1/Fs and the effect on the vehicle achieved velocity of an error in the estimation of the stage structure factor is reflected in the logarithm of the mass ratio of the stage. Figure 6.7 shows several examples of a vehicle model required to achieve 12000m/sec of ideal velocity.

The specific impulse, Isp and structure factor, Fs, of the stages are as shown while the mass allocation between stages was optimised for minimum GLOW. The table columns are: [1] MS, the stage mass, kg; [2] Fs, the stage structure factor; [3] Isp, the stage engine specific impulse, sec; [4] Ms, the stage structure mass, kg; [5] Mp, the stage propellant mass, kg; [6] Mign, the vehicle mass at stage ignition, kg; [7] Mbo, the vehicle mass at stage burnout (or shutdown), kg; [8] R, the vehicle mass ratio due to the stage in question; [9] ln(R), the natural logarithm of the mass ratio, [10] ∆V, the velocity change due to the stage in question. Case 2 is the baseline vehicle but with the second stage structure mass unable to be manufactured to the planned mass of Case 1. The mass has increased from 7.44% to 9.00% of the planned mass. This is an increase of 21% in the structure mass of the second stage. Case 2 shows that the effect is to increase the lift-off mass of the vehicle from 104091.5kg to 104447.1kg an increase of 355.6kg and a decrease in the achieved velocity from 12000m/s to 11903m/s, a loss of 97m/sec. Case 3 is the same baseline vehicle as in Case 1 but now with the second stage engines under performing and only delivering a specific impulse of 419.1sec, a decrease of 9.9sec or only 2.3%.

100

Chapter 6 Alternative Vehicle Concepts

Thus it can be seen that the achieved velocity is more sensitive to variations in the specific impulse of the engines than to an inability to quite meet the planned structure factor.

Figure 6.7: Three cases indicating the sensitivity of the achieved velocity to variations in stage manufactured structure factor and the engine achieved specific impulse

6.8 SENSITIVITY OF ORBIT INSERTION DUE TO LAUNCH ERRORS The relationship between the orbit semi-major axis, the orbital height and the velocity are given by the Vis-viva equation [eg ref 6.15], viz:

2 V µ ( −⋅= /1/2 ar ) ...... Eqn 6.4 where V = the velocity in the orbit, km, µ = the Earth’s gravitational constant = 3.9860044x105 km3/sec2, r = the radius in the orbit, km, a = the semi-major axis of the orbit, km and the semi-major axis

a = (rp+ra)/2.0 = (re+hp+ra)/2.0 where

rp = the radius of the perigee, km,

ra = the radius of the apogee, km,

hp = the height of the perigee above the Earth’s surface, km, and

re = the radius of the Earth, km.

101

Chapter 6 Alternative Vehicle Concepts

For an Ariane GTO, the perigee height is normally

hp = 200 km and

ra = 42164.5 km, the geostationary height and

re = 6378.14 km.

The above constants are taken from [Ref 6.16] Substituting these values in equation 6.4, we get

Vp = the velocity at GTO perigee = 10238.9 m/s.

For Ariane 5, Arianespace gives [Ref 6.5] a 1-sigma injection accuracy of 40km in the semi-major axis (a). As injection occurs near the perigee, a position error and a velocity error which affect the apogee height are represented. For Ariane 4LP, the user’s manual [Ref 6.2] gives a 1-sigma injection accuracy of 26 km in a, 1.0 km in hp and 52 km in ha. Although these values are applicable to the AR44LP, they differ very little from one Ariane 4 launch vehicle version to another. From the calculations above, the 1σ 52km value of dha translates to a 0.85 m/sec 1σ value of dVp, the corresponding perigee velocity error. Using the STAGEX spreadsheet and keeping the stage dry masses constant, an additional amount of propellant can be added to the third stage to determine the sensitivity of the ideal velocity to a propellant loading error. A propellant loading error of an additional 10.1 kg of propellant in the third stage or 20.3 kg in the first stage is sufficient to increase the Delta-V by 0.85 m/sec and the ha by the specified 1σ orbit insertion error. The propellant loading errors are only one source of error and all must be taken into account when specifying the required ideal velocity from the launch vehicle. Consequently the specified ideal velocity requirement includes an allowance of about 300 m/sec to account for the estimated errors to the 3σ level. If the launch vehicle performs nominally all of the reserve propellant will be unused while if the vehicle under performs at the 3σ level of the expected errors, all the reserve propellant will be utilised. Under actual flight conditions the flight computer in the vehicle equipment bay will monitor the velocity and shut down the engines when the correct velocity is reached.

102 Chapter 7 Trajectory Optimisation by Evolutionary Algorithm

7.1 TRAJECTORY SHAPING Previous chapters have considered the velocity budget of an Ariane 44L equivalent launch vehicle with primary mission to launch geostationary spacecraft equal to the Ariane vehicle. However differing designs such as vehicle acceleration and cross sectional area will affect such velocity budget components as gravity loss and drag loss respectively. Launch trajectory strategy and shaping will affect overall velocity requirements. Additionally, and not considered previously, there may be constraints on the trajectory such as the fallout zones of the fairing, first and second stages which may need to be controlled due to position of down range populations or physical assets which are to be avoided. Stage recovery and tracking may also be facilitated by a judicious push or pull of the fallout, by out of plane steering of the launch vehicle or by accepting slightly non- optimal trajectory. These factors can be adjusted by choice of stage size giving more or less performance to each stage while maintaining the full launch capability. It becomes obvious then that the launch system should be considered as a complete system. The location of the launch site, the characteristics of the launch corridor and operational factors should all be taken into account.

7.2 COMPUTER PROGRAM In order to more accurately model the velocity budget of the launch vehicle a computer program (TRAJ2DF) was written to integrate the equations of motion of the launch vehicle. The program was specifically written to model a three stage launch vehicle but may also accommodate a two stage vehicle if the appropriate parameters for the third stage are set to zero. The program included as variables the stage masses of the vehicle and a number of variables related to the flight profile. The program as written was a 2-D representation of the flight with a circular rotating Earth modeled.

7.2.1 Program Modes The program can operate in two modes: flight mode or optimisation mode. In the flight mode the input parameters are used to model the vehicle and trajectory and the mission is flown as specified to determine various output parameters including the final orbit and drop zones of the first and second stages and the fairing. In the optimisation mode a number of variables are given a range in which to vary and the program attempts to optimise (minimise) the vehicle lift-off weight under the constraint of a number of variables. The variables that can be constrained include the final orbit parameters perigee height (Hp) and apogee radius (Ra) while the variables to be determined as a result of the optimisation include the stage masses (M1, M2, M3), the vertical rise time (S1VRT), the initial first stage pitch angle (S1POA), the S1/S2 interstage coast time (S12ICT), the initial and final pitch angles for the second stage (S2IPA, S2FPA), the S2/S3 interstage coast time (S23ICT) and the initial and final pitch angles for the third stage (S3IPA, S3FPA). Minor modifications to the program would allow the drop zones of the stages and the fairing to be specified as constraints so that a region of down range launch corridor can

103

Chapter 7 Trajectory Optimisation either be avoided or specified as a drop zone. Many other constraints can also be added as required. These modifications will be discussed later in the section discussing future developments of the program.

7.2.2 Flight Profile Model The flight profile is modeled with a series of phases as shown in figure 7.1. The flight begins with a short period of vertical rise (S1VRT) from the launch pad followed by a pitch over to a maximum angle of attack. The angle of attack must be held to a small value otherwise aerodynamic bending loads can destroy the vehicle. Ariane-4 is limited to a four degree angle of attack before it exceeds ultimate loads. TRAJ2DF allows the maximum angle of attack to be specified as an input. These considerations point out a further series of optimisations that can be applied to a launch vehicle. A shorter fatter vehicle will be stiffer structurally allowing a faster pitch over and easier access to low earth orbits (LEO). This is at the expense of a larger diameter and hence higher aerodynamic drag and a greater possibility of sloshing in the fuel tanks. Furthermore, the shorter vehicle may require more difficult and expensive tank fabrication methods such as toroidal tanks. The shorter length may also compromise the command authority of the thrust-vectoring engines. Larger diameter tanks could also compromise their ability to be transported during manufacture and launch preparations. A design choice balancing all of the above factors must be made. However no attempt is made to optimise these factors in this work. Structural considerations are regarded as being taken up in the stage structure factor used in the calculations.

Figure 7.1: Flight Profile with Phases, Events and Associated Parameters

Once the specified pitch angle has been reached TRAJ2DF holds the pitch constant until the flight path angle equals the pitch angle (S1POA), i.e. the pitch angle is aligned with the velocity vector. The vehicle then follows a zero angle of attack (gravity turn; Ref: 7.2, pg 69) pitch program keeping the pitch aligned with the flight path angle. In the TRAJ2DF modeling, the gravity turn is maintained throughout the remainder of the first stage flight. As the flight dynamic pressure decreases with altitude, real launch vehicles may adopt increasing of attack between the thrust vector and the flight path but which are still

104

Chapter 7 Trajectory Optimisation constrained by the bending forces being applied at the time [Ref 7.6, pg72]. Outside the sensible atmosphere large angles of attack may be applied. TRAJ2DF considers this to occur from the end of the first stage powered flight. There is provision made for a period of interstage coast between the first and second stages (S12ICT) with the period specified as an input. This coast is usually fairly short, of say five seconds duration, and is usually only for stabilisation of the vehicle after staging and to allow the adoption of the initial pitch angle for second stage powered flight. Once second stage powered flight begins the vehicle follows a number of possible pitch programs. The two pitch programs currently implemented are zero angle of attack (gravity turn or tangential steering) and linear tangent steering. Due to the modular nature of the software more pitch programs such as polynomial programs may be added as desired. Linear tangent steering [Refs: 7.1; 7.2, pg 71] is designed to obtain the maximum δV for the minimum propellant expenditure during the period of the steering phase. It derives its name from the relationship below:

tan θt = tan θ0 +At ------Eqn 7.1

Where θt is the local pitch angle at time t into the steering phase, θ0 is the local pitch angle at the beginning of steering and A is a constant. Knowing the length of the steering phase, T, (i.e. the stage burn time) we have:

tan θT = tan θ0 + AT

Where θT is the pitch angle at the end of steering

A = (tan θT - tan θ0)/T ------Eqn 7.2

Or in the small angle approximation, the pitch rate is a constant

A= (θT - θ0)/T ------Eqn 7.3

dθ = AT ------Eqn 7.4

The local pitch angle can start at a large value such as 60 degrees at the start of second stage burn and end at say 30 degrees at the end of second stage burn so that the pitch rate is normally negative. At orbit insertion, at the end of third stage burn, for a circular orbit say, the local flight path angle must be zero but the pitch angle may not be zero and may even be negative, cancelling out the last of the vertical velocity. At the end of second stage burn a period of interstage (S23ICT) coast may follow. This can be zero or any length (time) and is specified as input to the program. The third stage then begins powered flight and again is subject to a number of possible pitch programs (once again, only two are currently implemented).

7.2.3 Trajectory Integration. Reference Frames: The model utilised in TRAJ2DF assumes a circular rotating Earth. To overcome the problem of relative velocities within the atmosphere, the reference frame is that of the launch site during the first stage burn. This allows aerodynamic drag to be calculated relative to the local frame of reference. After first stage burn1 and staging the Earth’s rotational velocity is added vectorially to the vehicle’s burnout velocity vector to change to

1 Approximately two minutes into the flight (or two minutes MEL = Mission Elapsed Time))

105

Chapter 7 Trajectory Optimisation an inertial frame of reference [Ref 7.3]. Launch site latitude and launch azimuth are read from the input file to calculate the component of Earth rotational velocity to be added. The Gravitational Force Model used during all of the flight is a simple Earth centered inverse square law with radius recalculated at each integration step. Again due to the modular nature of the program this model can be changed but for the mission concept planning uses of the 2-D program the simple gravitational model is deemed to be sufficient. Aerodynamic Drag: The Aerodynamic drag is implemented as the well known drag equation [Ref: 7.2, pg 68], viz:

D = Cd⋅Q⋅Aref ------Eqn 7.5 2 Q = ½⋅ρ⋅V ------Eqn 7.6 where D = the drag force, newtons

Cd = the dimensionless drag coefficient A = the reference cross-sectional area of the vehicle, m2 Q = the flight dynamic pressure, kgf/m2 ρ = the local atmospheric density, kg/m3 V = the local velocity of the vehicle, m/sec

Figure 7.2: Redstone drag coefficient curve from reference 7.4 (top) and the derived curve used in TRAJ2D

106

Chapter 7 Trajectory Optimisation

The drag coefficient, Cd, used in TRAJ2DF [figure 7.2] is derived from the curve given in Ref 7.4 for the Redstone vehicle. The curve is not read in from any data file but points on the curve were converted to a data array and permanently entered into the program. Due to the modular nature of the program provision was made for other data arrays representing other drag curves to be utilised and chosen as required with very little modification to the code. From the small number of data points entered into the program, a spline curve fitting subroutine interpolates the points to arrive at a value of Cd for any given velocity. Re-Entry Drag: The re-entry drag on the first and second stages and the fairing is ignored. Hence the fallout zones calculated in this version of the program are only indicative of the results that can be achieved. Besides the aerodynamic drag no other aerodynamic forces such as lift or wind gusts are considered in the program.

Engine Thrust: The thrust of the engines in the various stages can be specified as input either as a vacuum thrust or as a sea level thrust. From one the other is calculated. During the flight provision is made for the variation in the thrust due to the effect of decreasing atmospheric pressure with altitude. The thrust is calculated from equation 7.7 [Ref 7.5]

& e −+= aee )p(pA/gVWF ------Equ 7.7 where F = the thrust force, newtons

W& = the mass flow rate, kg/sec

Ve = the engine exhaust velocity, m/sec g = the acceleration due to gravity, m/sec2 2 Ae = the nozzle exit area, m 2 pe = the exhaust pressure at the nozzle exit, newton/m 2 pa = the ambient pressure (at the nozzle exit), newton/m

The first term on the right hand side is the momentum force while the second term is the pressure force. The pressure force, hence the thrust, is reduced by the atmospheric back pressure, pa, acting against the exhaust exit pressure, pe. As an aside, for the reader’s benefit, it can be seen that as the nozzle expansion ratio, ε, is increased to increase Ve, Ae increases and pe decreases with an overall increase in F in vacuum. However in the atmosphere Aepa increases with increasing ε. Therefore there is an optimum ε that should be chosen for first stage engines operating in the atmosphere. This optimum will depend on the time the engine spends at various levels in the atmosphere hence will depend on the acceleration and flight profile. Additionally the weight of the extended nozzle to provide increased ε must be taken into account when optimising the engine/vehicle combination. TRAJ2DF does not optimise the nozzle expansion ratio as engine mass models and flow separation caused by the overexpansion within the atmosphere are beyond the level of complexity of this version of the program. A value for ε derived from that used in historical engines is entered as data in the input file. Integration Method: The integration method used to determine the trajectory is a simple Euler method with adjustable time step length possible for each flight phase. This method was chosen not only for ease of programming but also for speed of computation. The

107

Chapter 7 Trajectory Optimisation modular method of construction of the program means that an alternative integration method can be easily implemented if required. Test runs of the program with varying time steps in flight mode indicated that 10-2 second is a suitable time step for first stage flight phases where the velocity achieved is only about 3000m/sec. While for second and third stage flight 10-3 sec is suitable.

7.2.4 Optimisation The second program mode is optimisation mode. The method used is the same as that for ROKOPT, i.e. the evolutionary algorithm by Ray and Sarker [Ref 5.4]. Instead of utilising a fixed deltaV as a constraint as in ROKOPT, the trajectory is now flown to obtain the required orbit as a constraint. The perigee height and apogee radius2 are specified to be observed within given tolerances. Both the values and tolerances of the perigee and apogee are input from the data file. The various flight parameters as described above are also input as data along with the range of the variables. Initial guesses of the stage masses obtained from STAGEX (Appendix C) are also input along with their allowable range. The program minimises the GLOW as the objective function and determines the values of the flight profile parameters that deliver the correct orbit and the minimum GLOW. The deltaV delivered by the minimum stack size found now more accurately reflects the velocity requirement under the design and flight constraints imposed by the designer. Appendix-B1 lists the input file for TRAJ2DF while Appendix-B2 lists an output file in flight mode for a representative vehicle. Appendix-C lists a representative optimisation mode output file. In optimisation mode there is also available a history file “HISTORY.OUT” that lists the values of the trial solutions for each population of each generation of each run of the program. This file is utilised in diagnosis of the progress of the convergence of the evolutionary algorithm. It can be several gigabytes in size and must be read by a special file editor which will handle large files. An example is not included as an appendix in this thesis. The generation of this file can be switched on or off by means of a software switch in the input file.

7.2.5 Performance of the Program The flight mode of the program takes only several seconds to perform the trajectory integration for the entire three stage vehicle, the first and second stage and fairing re-entry trajectories and to produce an output file reporting the vehicle design, launch trajectory and final orbit achieved. This may seem to be quite a reasonable time lapse. However in optimisation mode each member of the trial population must include a trajectory integration to obtain the values of the orbit perigee and apogee constraints. As mentioned in chapter 5 it is normally considered sufficient to use ten trial solutions for each variable. In the current problem there are at least ten variables and it would be desirable to add others as constraints, e.g. stage fallout ranges (as discussed earlier). This implies a trial population of 100 solutions that must be integrated. Examination of the history file generated by trial runs of the program indicated that a population of ten per variable, 80 members, was insufficient to obtain convergence and

2 Perigee height and Apogee radius are used as constraints because at perigee one is interested in the height above the earth’s surface and atmosphere whereas at apogee it is the geosynchronous radius that is of interest as a reference radius as the primary launch mission is into a geosynchronous transfer orbit.

108

Chapter 7 Trajectory Optimisation that at least fifty full generations of evolution and a population of 130 was required to reach convergence. Even this was only just sufficient to converge. A much more populous solution set would be desirable. Given the above situation the program took about 12 hours to complete a single run with a single set of optimisation parameters. Only two runs could then be achieved per day of elapsed time. The program was allowed to run for several months to generate several hundred runs from which the minimum GLOW was to be selected. Late in the period of the program development the computer was changed from a 3.0 GHz Intel Pentium 4 CPU based machine to a 2.3 GHz Intel Core 2 Quad CPU based machine. This resulted in an approximately three to four-fold increase in computational speed. With a population size of 150 and 120 generations it then became possible to carry out six to eight runs a day.

Table 7.1: Optimisation parameters for a 128 run optimisation of ALV-4 Parameter Variable Minimum Maximum Variables with range S1 mass, kg S1MASS 90000 110000 S2 mass, kg S2MASS 45000 60000 S3 mass, kg S3MASS 8000 18000 S1 pitch over angle, deg S1POA 8.0 15.0 S2 initial pitch angle, deg S2IPA 30.0 50.0 S2 final pitch angle, deg S2FPA 10.0 30.0 S3 initial pitch angle, deg S3IPA 5.0 10.0 S3 final pitch angle, deg S3FPA -2.0 5.0

Fixed value variables S1 Engine expansion ratio S1Eps 10.48 First stage vertical rise time, sec S1VRT 7.0 S12 interstage coast time, sec S12ICT 0.0 S23 interstage coast time, sec S23ICT 0.0

Objective Function to minimise GLOW

Constraints Perigee height ± tolerance, km Hp ± Hptol 200.0 ± 12.0 Apogee radius ± tolerance, km Ra ± Ratol 42165.5 ± 156.0

Optimisation Parameters Number of solutions NumSol 150 Number of generations NumGen 120 Probability of crossover ProbX 0.90, 0.92, 0.94 Probability of mutation ProbM 0.50, 0.52 ,0.54 Distribution index of crossover DIX 10.0, 15.0, 20.0 Distribution index of mutation DIM 10.0, 15.0, 20.0

109

Chapter 7 Trajectory Optimisation

Trial runs further indicated that operation was satisfactory except for the long run times involved. As a consequence it was not possible to generate more than a few representative optimisations for the preferred ALV-4 vehicle. Examination of the history file for the test runs seemed to indicate that the objective function is flat in variable space around the minimum values of the GLOW and the solutions for the trajectory variables vary considerably from run to run. Hence many more trial solutions are needed to effectively fill the manifold and determine the trajectory solution set corresponding to the minimum GLOW. The solutions found produce flyable trajectories however, as will be shown in the next chapter. It was found that the pitch-over of the vehicle during first stage operation takes a considerable fraction of the burn time of the first stage because of the angle of attack limitations. In order to start the pitch over as soon as possible vertical rise time was set to a value just sufficient to allow the vehicle to clear the launch tower. The range on this variable was set to zero to help speed up convergence of the program. Likewise the S2/S3 interstage coast time was set to zero along with its range. Table 7.1 shows the parameters used for a 128 run optimisation of the ALV-4 stage masses and trajectory parameters minimising the GLOW. From the output of the run the results were sorted in order of the GLOW obtained. The best 20 of these runs are shown in figure 7.3 along with some statistics of the stage masses corresponding to each GLOW value.

Figure 7.3: Spreadsheet showing best (lowest GLOW) 20 runs of a 128 run set of optimisations of the ALV-4. Statistics of the stage masses and trajectory parameters (the solutions) are also shown.

The lowest value of the objective function (GLOW) was found to be considerably higher than that found by the simpler assumptions of ROKOPT, ie 176,665kg against 143,456kg, a value of 33,209kg greater. The solutions for the stage masses and in particular the trajectory steering parameters have not settled to anywhere near a single value set. Furthermore some of the steering parameter solutions are sitting on the limits of the range for these variables indicating that for these particular solution sets a wider range of the variable would be desirable.

110

Chapter 7 Trajectory Optimisation

Test runs widening the limits of the variables, i.e. increasing the range, caused the program to not converge the population of each run even with a much larger number of generations. This would indicate that the size of the population used is too small and more members are needed to fill the variable space adequately. Geethaikrishnan et al recognise some of the same problems. In their paper [Ref 7.6] they describe the need for a Genetic Algorithm Guided Gradient Search (GAGGS) to optimise a space shuttle launch trajectory. They utilise the capability of genetic algorithms to explore the entire variable space and a gradient method to quickly converge to the local optimum starting from the best solution obtained by the genetic algorithm. Chapter 5 of this work mentions the need for this in ROKOPT and it would also appear to be a useful addition to the methods of TRAJ2DF. The trajectories found by TRAJ2DF provide a satisfactory flight however, as evidenced by the final orbit parameters of the perigee height, Hp, and the apogee radius, Ra. The greater gross lift-off weight (GLOW) reported by TRAJ2DF can be attributed to differences between assumed values of the specific impulses (Ivac) for the stages in ROKOPT and the calculated specific impulses in TRAJ2DF. TRAJ2DF uses the methods outlined in Huzel and Huang [Ref 7.5] to calculate engine performance (Ivac) from thermodynamic properties of gases whereas ROKOPT uses estimated figures from historic engines. In particular the difference between Ivac and Isl is estimated by ROKOPT whereas TRAJ2DF does a calculation for the difference and adjusts the value of Isp between Isl and Ivac with increasing altitude. Utilising the solution set found for the minimum GLOW of figure 7.3, the ideal velocity of the two models is slightly different. The ROKOPT value comes from the AR-44L vehicle as calculated in Chapter 6 while the TRAJ2DF value is that achieved by calculation. The major differences between the solutions of the two programs are listed below in the Table 7.2

Table 7.2: Major differences between ROKOPT and TRAJ2DF Parameter ROKOPT TRAJ2DF

S3 Ivac, sec 445.1 445.6

S2 Ivac, sec 430.0 421.6

S1 Ivac, sec 430.0 397.4

Vtot, (Ideal Velocity), m/sec 11945.5 11877.3 Fairing jettison height, m End S2 burn 110,000

Substituting TRAJ2DF calculated Ivac and Vtot values back into STAGEX/ROKOPT the following comparison of table 7.3 can be made

Table 7.3: Comparison of ROKOPT and TRAJ2DF results using TRAJ2DF calculated engine performance data Parameter ROKOPT TRAJ2DF S3 Mass, kg 14484 12770 S2 Mass, kg 59281 53732 S1 Mass, kg 95523 103965 GLOW, kg 176025 176665

While the lift-off weights are comparable, differing by only 640kg (0.36%), the stage masses vary considerably, again indicating that the objective function is flat near the minimum.

111

Chapter 7 Trajectory Optimisation

7.3 PROCESSING SPEED LIMITS THE RESULTS FROM THE SOFTWARE Even with the increase in speed made possible by the Intel Core 2 Quad CPU the processing speed was too slow to allow further effective testing and diagnosis of the program and analysis of the results. It was certainly too slow to add further variables to the solutions. It was evident then that the program would need to be implemented on a parallel processor so further development of the program was halted at this point. Plans are in hand to implement the program on the 1100 CPU cluster at ANU3. This machine has a Linux operating system and the FORTRAN and C++ development environments are of course totally different to the outdated and now unsupported COMPAQ Visual FORTRAN 6.6 used for this work so far. It will therefore be necessary to rewrite the software in a form to allow each individual run to be submitted to a separate CPU and the results collected as they are completed. In this way the result from the first run will not be available for several hours but results from each successive run will come in steadily every few seconds after the first. Thus several hundred runs with a much larger population size evolving over many more generations will be computed each day of elapsed time. This will allow easier analysis of the program performance and of the optimisation results. In addition gradient methods will be added to aid in the convergence to the local optimum from a starting point given by the best result of the genetic algorithm. Even at the current stage of development TRAJ2DF represents a useful tool for doing conceptual studies of launch vehicle sizing and trajectory parameters along with stage fallout ranges. The TRAJ2DF results of figure 7.3 will be used in Chapter 8 where launch tracks and fallout will be examined further with emphasis on the Australian situation and launch system concept decision making.

3 ANU – Australian National University, in Canberra, ACT

112 Chapter 8 Launch Site and Corridors

8.1 INTRODUCTION When considering the design of a space launch capability the launch site is arguably the most important component of the overall system. The latitude of the launch site along with available launch azimuths and corridors will determine either the orbits available for direct injection without the requirement for a plane change or the size of the plane change, where necessary. This will affect the size of the launch vehicle required as the primary mission is to the important equatorial geostationary orbit. An increase in plane change will increase the velocity increment required at apogee with an attendant increase in the size of the apogee boost motor which implies an increase in launch payload mass. While the above effect is normally the one considered as the driving factor in the choice of launch site, with the desire being to be as close to the equator as possible, by far the more important consideration is the economic geography of the launch site. It is at the launch site that most of the activity occurs during a launch campaign and indeed many of the other activities associated with a space launch system can also be usefully located at the launch site. As was discussed in chapter three, it is the operational phase of a space launch that incurs the ongoing lifecycle costs and indeed, has the potential to provide an ongoing income. The placement of a launch site then is governed by two constraints: A) Possible locations B) Desirable locations A) Possible Locations are determined primarily by range safety and availability of sufficient suitable land. Whether a given parcel of land is suitable is determined by safety considerations with respect to its distance from population centres and/or other physical assets and the location of launch corridors and trajectories that may overfly populated areas. B) Desirable Locations are determined by their proximity to the equator but more importantly proximity to manufacturing centres, suitable existing transport and other facilities such as accommodation, electricity and water supply, sewerage and communications. The initial cost, ongoing maintenance and operation of such facilities should be minimised by connecting to existing networks where possible. Both of these two constraints must be satisfied. The first must be absolutely satisfied and the second should be optimised for minimum lifecycle cost.

8.2 ALV-4 LAUNCH CORRIDORS Using the results of the run producing the lowest GLOW in chapter 7, the ALV-4 vehicle was “flown” by TRAJ2DF and the flight events tabulated below in table 8.1. The data listed represents a nominal flight and in the following discussion the small variations in the figures due to off-nominal performance1 should be borne in mind. These variations are to be determined once a vehicle design is accomplished and the appropriate performance margins established.

1 But still resulting in the desired orbital outcome.

113

Chapter 8 Launch Site and Corridors

Table 8.1: Flight Data and Event Sequence for ALV-4 Launch Vehicle Nominal Flight using Optimised Trajectory of Chapter 7.

Case Study: ALV-4 Launch Vehicle FLIGHT SUMMARY – Nominal Mission Event Listing FIRST STAGE (Launch Site Relative Frame) Start Pitch Over Altitude 102.4 m Start Pitch Over Velocity 29.6 m/sec End Pitch Over Altitude 4914.7 m End Pitch Over Velocity 230.5 m/sec End Pitch Over Downrange 385.8 m Max Q Time 60.2 sec Max Q Altitude 8748.1 m, (28701 feet) Max Q Velocity 323.5 m/sec Max Q Mach No 1.1 Max Q Downrange Distance 1199.6 m S1 Burn Out Time 143.5 sec S1 Burn Out Altitude 64233.6 m, (210740 feet) S1 Burn Out Velocity 1627.4 m/sec S1 Burn Out Mach No 5.3 S1 Burn Out Downrange Distance 42227.6 m S1 Fallout Time 432.8 sec S1 Fallout Range 377.2 km SECOND STAGE (Inertial Frame, No Rentry Drag) S2 Shutdown Time 489.0 sec S2 Shutdown Altitude 221076.5 m, (725316 feet) S2 Shutdown Velocity 6076.4 m/sec S2 Shutdown Downrange Distance 1176.3 km S2 Fallout Time 870.4 sec S2 Fallout Range 3420.5 km FAIRING (Inertial Frame, No Rentry Drag) Fairing Fallout Time 478.4 sec Fairing Fallout Range 680.9 km ORBIT ATTAINED (Perigee Height ≥ 100km, Inertial Frame) Mission Elapsed Time 661.2 sec Mission Altitude 211701.0 m Mission Downrange Distance 2319.1 km THIRD STAGE (Inertial Frame) S3 Ignition Time 489.0 sec S3 Ignition Altitude 221076.5 m S3 Ignition Velocity 6076.4 m/sec S3 Ignition Downrange Distance 1176.3 km S3 Shutdown Time 817.9 sec S3 Shutdown Altitude 215969.2 m S3 Shutdown Velocity 10225.5 m/sec S3 Shutdown Downrange Distance 3668.7 km S3 Shutdown Perigee Height 208.2 km S3 Shutdown Apogee Radius 42217.1 km S3 Perigee Time N/A sec S3 Perigee Downrange Distance 3193.3 km

Notes: Perigee time is unavailable as at 3rd stage shutdown the vehicle is past perigee in the ascending part of the transfer orbit. “Orbit Attained” refers to point where perigee has been raised to above 100km.

114

Chapter 8 Launch Site and Corridors

8.2.1 Locating the Eastern Launch Site and Corridor As mentioned in Chapter 3, in the absence of a promontory in the south east corner of the continent at a suitably low latitude with launch sectors clear of inhabited areas and protected assets, two launch sites are required: one for launches to the East for low inclination orbits and one to the South for polar and sun-synchronous orbits. As the primary mission is to be to the equatorial geostationary transfer orbit (GTO), the apogee of the orbit must be on or near the equator. This implies that perigee of the transfer orbit must also be on or near the equator. For the ALV-4 launch vehicle a direct injection into geostationary transfer orbit (i.e. no-interstage coast) third stage shutdown and orbit injection occurs approximately 3668km downrange: this point must be near the equator. Figure 8.1 show the situation for a direct injection ALV-4 launch from a site in northern Queensland, viz, Cairns. The red marker “P” represent the position of perigee at S3 (nominal) shutdown, the red marker “O” represent the position at which orbit is achieved defined by perigee raising above 100km. The partially obscured red marker “L” is the launch site. In order to place perigee on the equator the launch azimuth must be such that the GTO inclination is approximately 45 degrees. This incurs velocity penalties in plane change at apogee, a larger apogee motor, smaller payload or the requirement for a larger launch vehicle. Furthermore, the trajectory and stage fallout is over or near populated islands of the Coral Sea, Papua New Guinea, Bougainville and the Solomon Islands. Moving the launch site further North of Cairns towards the equator while improving (decreasing) the GTO inclination does not overcome the problem of over flight of inhabited islands for any launch azimuth and the accessibility of a putative launch site is only made worse. One of the major failings of the 1986 Cape York International Spaceport proposal [Refs 1.9 – 1.14] was the requirement to construct long distances of road and/or rail infrastructure along with the many bridges required to span the numerous creeks between Cairns and the Cape York Peninsula to enable access to the launch site. Moving further south the situation is made worse as the GTO inclination is greater and at the latitude of Brisbane a direct injection strategy cannot reach the equator before S3 shutdown. The lower latitudes do not improve the ground track of the launch trajectory with respect to passage over inhabited or sensitive areas. An alternative strategy is therefore required. Launch Strategy for Eastern Launch Site (PLR = Pacific Launch Range): In order to satisfy the range safety requirements it is necessary to move south of Northern Queensland and look for locations where a trajectory will not over fly populated areas and where there is sufficient land for launch safety. At about the latitude of Gladstone in Queensland the number of Pacific islands to the east begins to decrease and a clear flight is possible. It is for this reason that a relatively recent launch site proposal [Refs 8.1, 8.2] chose Hummock Hill Island (Lat 24.0°S) just off the coast as their preferred location. However the presence of the Great Barrier Reef and several Australian tourist islands make this an environmentally and politically sensitive area. An examination of the map of the eastern coastline of Australia reveals that there is very little land that has not been developed and populated. However there is one small parcel that solves the problem of range safety, of placing the perigee on the equator and is desirable in operational terms in that it is close to industrial centres. It comes with a small penalty however. The fortuitous site is the naval gunnery range on Beecroft peninsula. This site is 12.5 km from Jervis Bay which also hosts the HMAS Albatross naval air station. It is close to the industrial centres of Sydney, Wollongong and Melbourne and the administrative centre and seat of the Australian Federal Government in Canberra ACT.

115

Chapter 8 Launch Site and Corridors

116

Chapter 8 Launch Site and Corridors

Figure 8.2 shows an overview and detailed map of the area while Figure 8.3 shows the strategy for launch operations from Beecroft Peninsula (the proposed PLR: Pacific Launch Range). The putative site is located on a peninsula of land owned by the , several kilometers from population centres and surrounded by several gun emplacements utilised for naval gunnery practice. Danger and restricted areas for air and sea are already in place for this purpose. The site is close to major transport routes to major industrial centres. Permanent housing and temporary accommodation for staff is available in nearby towns eliminating the need for construction of residences for staff needed during a launch campaign. An ALV-4 launch to GTO commences with a short vertical rise of seven seconds followed by the pitch over to commence the gravity turn. The flight proceeds over empty ocean clear of populated areas and passes New Zealand some 217km on the right (to the south). First stage burnout and fallout into the Tasman Sea occurs well before New Zealand is reached as does fairing fallout (see also table 8.1). Second stage shutdown and third stage ignition also occur over the Taman sea well before the flight passes the northern tip of New Zealand. The Kermadec island group, Raoul, Macauley, Cheeseman and Curtis are volcanic islands of New Zealand sovereignty. They lie on the Kermadec oceanic ridge some 2900 kilometers downrange. The islands are virtually uninhabited but with a manned volcanic and conservation observatory on Raoul island. The trajectory passes directly overhead the Kermadec group which would make an ideal site for a downrange tracking station (with the permission and cooperation of the New Zealand Government). Orbit insertion2 into a low earth parking orbit occurs north of the northern tip of the north island of New Zealand whence the third stage engine is shutdown. The vehicle now coasts 3 in a parking orbit of dimensions ha=300km, hp=100km. See Ref 8.3 for comparison of direct injection and parking orbit strategies. Second stage fallout into empty ocean occurs some 450 kilometers past the Kermandec group The vehicle continues coasting towards the equator passing directly over the island of Tahiti4 and other islands of French Polynesia. Approaching the equator the third stage engine is re-ignited. A small yaw to the right (i.e. to the south) is set in, enabling the third stage to take out several degrees of orbit inclination5. The third stage continues to burn until apogee is raised to the required geosynchronous height whereupon it is shutdown (after passing somewhat north of the equator). While the proposed PLR appears to be suitable for GTO and other low inclination launches, it comes with a penalty. The putative launch site is at a higher southerly latitude (35.05°S) than would be achieved by a site at, say, Gladstone. As has been mentioned earlier this means a higher inclination transfer orbit requiring a larger ABM for plane change and circularisation at Apogee. For the same useful payload into drift orbit this means a larger launch payload and a larger launch vehicle to lift it. As this is a clean sheet system design, a larger launch vehicle can be provided at some cost but by far the most important matter is the lifecycle operational savings due to the proximity of the launch operations to industrial centres (See Chapter 3 – System Considerations).

2 Once again attaining orbit is defined by the perigee height raising above 100km 3 This is the ALV-4 direct insertion trajectory as previously calculated. In the actual case the orbit would be adjusted slightly to result in a probably circular parking orbit with hp of about 200km. 4 It is almost impossible to avoid over-flying Tahiti for any due east launch from the east coast of Australia! 5 As the specific impulse of the third stage is much greater than that of the ABM (Apogee Boost Motor), it is efficient to take out several degrees of inclination with the third stage rather than to leave the full plane change to the ABM at apogee.

117

Chapter 8 Launch Site and Corridors

Figure 8.2: Location of eastern launch site at Beecroft Peninsula. Top: Overview map of South-Eastern Australia with location of Beecroft Peninsula. Bottom: Map of area surrounding Beecroft peninsula showing distances from launch site to nearby towns

118

Chapter 8 Launch Site and Corridors

119

Chapter 8 Launch Site and Corridors

8.2.2 Locating the Southern Launch Site and Corridor The location of the Southern Launch site is less of a problem than that of the Eastern Launch site as it is not constrained by the requirement to be as close to the equator as possible. In addition the South coast of Australia is much less fertile and hence less populated and than that of the Eastern seaboard. The main requirement is to be as close as possible to population centres for access to services and amenities. There is however a secondary consideration and that is a clear launch azimuth for International Space Station (ISS) resupply missions. Polar and sun synchronous launches require an almost due south launch whereas ISS orbit resupply missions an orbit inclination of approximately 51.6 degrees, requiring a south easterly launch azimuth for direct injection launches. It may be possible to carry out some polar or sun-synchronous launches from the Beecroft Peninsula by tracking South West along the New South Wales and Victorian coasts (depending upon detailed range safety requirements). However it is probably not possible to carry out direct injection ISS resupply missions from there as the trajectory would directly over the South island of New Zealand. A second launch site is therefore required. Launch Strategy for Southern Launch Site (SLR = Southern Launch Range): Examination of the map of the southern coastline of Australia reveals a number of possible locations for a launch site however one appears to be superior to all others. An area to the East of Esperance in the state of Western Australia called Point Malcolm gives a clear launch to the South East for ISS resupply missions and a clear launch to the south across the southern ocean with no intervening land masses before Antarctica. Figure 8.4 shows the overview and detail maps of the proposed launch site at Point Malcolm. The site is 172km east of Esperance on the Western Australian south coast. And 23km from the small settlement of Israelite Bay to the north east. The site has clear launch azimuths from about 45degreees through to about 135 degrees suitable for sun- synchronous, polar and ISS re-supply launches. The site lies on the eastern boundary of a conservation region and is 52 km from the eastern edge of the nearest farmland. This site is not as suitable for communications (of all forms) purposes as that of Beecroft peninsula however major rail and road connections exist with the rest of the nation from Esperance. A dirt road connects Esperance to Point Malcolm and on to Israelite Bay. Figure 8.5 shows the approximate ground trace for an ALV-4 launch into a 97 deg inclination sun-synchronous orbit. S1 fallout and fairing fallout both occur over the southern ocean. However S2 fallout occurs close to the Antarctic coast line at the Australian research station of Casey base. This is a prime example of the possible need for further constraints in the TRAJ2DF program to modify the stage masses and cause the S2 fallout to move further back from Casey. The bases at Casey, Davis and Mawson are all ideal locations for tracking stations for Sun- synchronous and polar launches although tracking and telemetry reception earlier in the flight would need to be accomplished shipboard from a vessel located in the southern ocean. For direct injection ISS re-supply missions, a south easterly launch from Point Malcolm would pass south of Tasmania, clear of all inhabited areas. The ground track continues south of New Zealand across the southern Pacific ocean and crosses hhe American continents at Panama and then on to Santo Domingo. Tracking stations at Ceduna on the

120

Chapter 8 Launch Site and Corridors southern coast of the Australian mainland and/or southern Tasmania would appear to provide sufficient coverage during the launch phase. Figure 8.6 shows an ISS resupply launch ground track and trajectory calculated for an HLV launch vehicle (original Ariane-4L analogue as modeled in this thesis). The main launch events and mission elapsed times (MEL) are shown.

Figure 8.4: Location of southern launch site at Point Malcolm. Top: Overview map of South-Western Australia with location of Point Malcolm. Bottom: Map of area between Esperance and Point Malcolm showing distances from launch site to nearby locations.

121

Chapter 8 Launch Site and Corridors

122

Chapter 8 Launch Site and Corridors

123

Chapter 8 Launch Site and Corridors

8.3 SUMMARY Approximately fifty percent of all launches are to the equatorial geostationary orbit. For these missions a near equatorial launch site appears to be out of the question because of the proximity of inhabited islands to the east of the Australian continent. Environmental complications due to the presence of the Great Barrier Reef and the availability of sufficient suitable land are further difficulties. By accepting the need to carry out a larger plane change for low inclination orbits a suitable location can be found at the Beecroft peninsula in New South Wales. The proposed site is at the range-head of the Royal Australian Navy gunnery range. A low earth parking orbit and a third stage restart must be accepted. Tracking and telemetry reception can be accomplished by stations located on the islands of the New Zealand Kermadec island group and shipboard on vessels stationed in the Pacific Ocean. In order to provide the same payload capabilities as for a launch from a lower latitude site the launch vehicle (LV) must be designed to be larger than would otherwise be required. As this is a study for a clean sheet system design the LV can be designed to be that much larger at some cost but the financial lifecycle advantages of a launch site close to industrial areas with short lines of communications greatly outweighs the extra cost of the LV. Operational costs dominate the lifecycle costs of a launch vehicle system. For Polar and sun-synchronous missions, if not possible from Beecroft Peninsula, and ISS re-supply missions a launch site at Point Malcolm to the east of Esperance on the south coast of Western Australia gives clear launches between Australia and the Antarctic continent. The location of a Point Malcolm site is not as convenient for communications and establishment of facilities as is the location of the Beecroft Peninsula site; however it is considerably more advantageous than that of the European launch site at Kourou in French Guiana where overseas shipping of materials and personnel from Europe must be employed. Tracking stations can be located on the Australian Antarctic Territory for Sun-synchronous and polar launches, while stations typically located at Tasmania and at Ceduna on the Australian mainland will serve for ISS re-supply launches.

ACKNOWLEDGEMENT The use of the “Google Earth” program and the KML (Keyhole Mark-up Language) to produce the maps and trajectories is acknowledged.

124 Chapter 9 Synthesis of System Design Concepts

9.1 INTRODUCTION A single Ariane-5 launch costs in the order of USD120 million [Ref 9.1]. This is made up primarily of labour costs and as such is a source of employment. With an estimated launch rate of between four and twelve a year the income from a mature space launch system would exceed the annual turnover of many of the primary industries that the Australian government currently spends considerable time and effort worrying about and propping up economically and politically. Decision makers only see the word “cost” and do not translate that to “income” which would apply if this nation were participating in the commercial space industry. While government support would be necessary in the initial development phase, a mature, revenue positive launch system, could be self supporting. London [Ref 9.2, p118] states “The failure of the US commercial launch industry would represent the loss of a national technical and defense treasure, many millions of dollars in commercial revenue and numerous jobs”. The same could apply to Australia. The sight of the many highly public “blow-ups” on the pad during the early days of the US space program is still remembered by the decision makers. The nearly flawless development of the British Blue Streak vehicle and its launches from this country do not seem to have had a similar reverse mental impact. The orderly development of the European Ariane series of launch vehicles as a commercial enterprise does not appear to have penetrated at all into the consciousness of politicians and bureaucrats. Consequently it appears that if Australia is ever to participate in the commercial space launch industry the initial impetus for the development of a launch system will fall to non-government organisations. There are a number of groups existing in Australia attempting to develop what they hope will become a national launch vehicle system. These groups come and go and their vehicle designs vary in size and in concept. There are air breathing hypersonic systems, solid propellant, liquid bipropellant and hybrid systems all under consideration. Although a considerable body of expertise and facilities still remains within DSTO from the Blue-Streak Woomera days this expertise is not being employed in the space arena and the current groups are mainly within universities, private research institutes or amateur organisations. The aim of this thesis therefore has been to examine a number of the major system design choices for a possible Australian space launch vehicle system. It was not intended to do a detailed design but rather to consider those elements for which choices must be made at the concept exploration phase of a project in order to save changes being made part way through the project. Historical data indicates that 70% of the entire life cycle costs of US Department of Defense weapon systems are fixed in the concept exploration phase of the project: see London in Wertz and Larson, [Ref 9.2, Chap 4]. Such decisions could compromise the continuation of a project if made incorrectly so it is important that they are made correctly from the start.

9.2 THE SYSTEM While United States launch vehicles such as the Delta, Atlas and Titan were based on military ballistic missile programs, the very successful European Ariane family was designed specifically for the commercial market to geostationary orbit. The heaviest lift vehicle of that family (before the Ariane-5) the Ariane 44L has been chosen as the archetype launch vehicle to reverse engineer for an Australian launch system.

125

Chapter 9 System Synthesis

9.2.1 Payload Capability The actual payload capability to be provided should be determined by examination of the market niches at the time of system development by carrying out a market survey. The AR44L is now out of Service and ESA relies on the Ariane 5 heavy lift vehicle with a payload capability of 10 tonnes into GTO and the Soyuz vehicle with payload capability of 3,150kg into GTO and from 2010 the lightweight launch vehicle, tailored for small to medium-sized payloads on missions to LEO and Sun-synchronous orbit. Vega will have a payload capability of 1,500kg into a 700km circular polar orbit [Ref 9.3, 9.4]. In comparison AR44L had a payload capability of 4,768kg to GTO [Ref 9.5]. Ariane-10-tonnes will be the standard version of the Ariane 5 launcher for the next few years but a successor is already being planned under the Ariane 5 Plus program. This version is set to increase its GTO lift capability to 12 tonnes [Refs 9.6, 9.7]. NASA is also looking to obtain a new launch vehicle to replace the Delta II for medium size payloads [Ref 9.8]. Payload requirements are continuously changing as the largest payloads get larger and smaller ones also get larger to fill the gap. A capability between the AR44L and the current AR5 would be a good place to start in order to capture future growth of present day Soyuz type payloads. Continuous expansion of capability is always to be kept in mind.

9.2.2 Mission As approximately fifty percent of all space missions are to geostationary orbit [Ref 9.9] the primary mission is to this orbit. Low earth orbit missions to low inclination, polar and sun- synchronous orbit for scientific and military purposes are also important and should also be catered for. ISS resupply missions require larger payloads but are also an important mission. Lunar and planetary missions are less common but should also be considered in system planning.

9.2.3 Launch Vehicle Concept

Propellant: The current trend is to move away from the highly toxic N2O4/Hydrazine propellant combination. This thesis has examined a number of propellant combinations in the Australian context and two would appear to be most favourable. They are LOX/Propane and LOX/Liquid Hydrogen. LOX/Propane is desirable for the high density of sub-cooled liquid propane which is temperature compatible with liquid oxygen as well as for its ready availability and low processing requirements. LOX/Liquid Hydrogen is desirable for its high performance and its clean exhaust products (Steam) but its low temperature and density are drawbacks. The low density requires larger fuel tanks but chapter six has shown that this is not an overwhelming drawback. Most current upper stages are LOX/LH2 for performance reasons so an Australian launch vehicle would also be likely to use this propellant in the third stage. For development and support reasons all stages should utilise common propellants. Therefore LOX/LH2 is to be recommended as the propellant of choice for all stages for the reasons of high performance, being a single technology to master and for its environmental desirability. As for pressurants: at the time of writing Central Petroleum have still not found commercial quantities of Helium in Australia. The Magee No.1 well showed 6.2% Helium third in quantity after Nitrogen and Methane in drilling in 1992 and the Mount Kitty area is expected to be better. The presence of the gas is still only prospective however.

126

Chapter 9 System Synthesis

Configuration: The configuration recommended is that of the tried and true three stage expendable launch vehicle. The architecture is that of the Ariane 1 to 4 family but without the strap-on boosters of the later models of the family. Rejecting the use of strap-ons reduces the number of components to manufacture, transport and stack but increases the size and weight of the core vehicle. Vehicle stage masses have been calculated for the ALV-2 and ALV-4 vehicles in Chapter 6. The assumption at that time was that the launch site was at CSG, Kourou. Adjusting for the latitude of the launch site to be at Beecroft Peninsula (35°S), by the method in chapter 3.3.2, an additional factor of 1.22 allows for the increased velocity for plane change. The following masses of the two preferred vehicles along with a larger vehicle equivalent to the present Ariane-5 vehicle are set out in Table 9.1.

Table 9.1: Vehicle Masses for Beecroft Peninsula Launch to Geostationary Transfer Orbit for AR44L and AR5Equivalent Payloads Vehicle ALV-2 ALV-4 “Ten Tonne” LOX/LH LOX/LH LOX/LH 3rd Stage 2 2 2 34598 kg 14079 kg 29528 kg LOX/C H LOX/LH LOX/LH 2nd Stage 3 8 2 2 122536 kg 40044 kg 83985 kg LOX/C H LOX/LH LOX/LH 1st Stage 3 8 2 2 133836 kg 113331 kg 237693 kg GLOW 298532 kg 175016 kg 367066 kg Engines: As explained in Chapter 3 the basic concept for an Australian launch vehicle should utilise turbo pump fed engines rather than pressure fed engines. The question arises however as to what engine cycle should be employed. Table 9.2 is taken directly from Ref 9.2 and sets out the trade off between gas generation cycle engines and more complex (staged combustion1) cycles such as are utilised by the Space Shuttle main engines. Table 9.2 leaves out the comparison with pressure fed-engines.

Table 9.2: Engine Cycle Comparison [Ref 9.8] Engine Cycle

Gas Generator Closed Cycle Vehicle weight ratio 1.25 1.0 Chamber pressure 800 – 1500 psi 2000 – 4000 psi 300 (LOX/RP-1) 320 (LOX/RP-1) Average flight Isp, sec 400 (LOX/LH2) 430 (LOX/LH2) Vehicle Complexity, Relative 1.0 1.6 Engine component count 2000 - 3000 5000 - 7000 Manufacturing complexity 1.0 1.75

1 Staged Combustion or closed cycle combustion uses main propellant to drive the turbopumps whose exhaust is fed back into the propellant stream and thence into the engine to complete combustion. There are a number of different versions of closed cycle combustion utilising the propellant in different cycle stages for different fractions of the auxiliary power. The Gas Generator cycle is different in that some propellant is utilised to drive the turbopumps at a lower pressure and temperature and at a more fuel rich mixture than in the main engine. The turbopump exhaust is then dumped overboard and provides very little effective thrust.

127

Chapter 9 System Synthesis

In June 1959 Theodore Von Karman observed [Ref 9.10]: It is my personal belief that the length of the period of attaining reasonable reliability in the development process could be essentially reduced if simple design were emphasized as a leading principle, even if we had to make some sacrifice in the quantitative measure of “efficiency.” Essential elements have to be designed as simply as possible, even if this means a reduction in quantitative efficiency and a certain increase of bulkiness and/or weight Or to put it more concisely: Take the performance hit and keep it simple. Table 9.2 demonstrates the large increase in part numbers and manufacturing difficulty for closed cycle combustion engines over gas generator engines. The trade off is in lower Isp and hence increased stage weight to achieve the same Delta-V. In fact Table 9.2 is pessimistic as the Ariane HM7B LOX/LH2 third stage engine which utilised a gas generator cycle produced Ivac efficiencies of 445.1 seconds at chamber pressures of only 35 bar (508 psi, ~3.5 MPa). The Space Shuttle staged combustion LOX/LH2 main engines produced an Ivac of 452.1sec only 7 seconds greater than the HM7B while running at chamber pressures of 189.6 bar (2747 psi, ~19.0 MPa) [Ref 9.11]. High chamber pressure does aid in reducing thrust-atmospheric losses in the lower atmosphere however. In the Australian context where the development of liquid propellant engines would be a new experience the gas generator cycle should be recommended. The low parts count and lower manufacturing complexity are the advantages here. Perhaps a chamber pressure at the upper end of the range, say 10MPa (~1500psi), could be attempted. The third stage engine at least should be capable of multiple restarts. It may be of advantage to provide that capability for the common first and second stage engines as well to enable them to be used as third stage engines in any future larger launch vehicle. A NASA Shuttle operations study found that simplicity is the key to reducing launch costs [Ref 9.12]. London [Ref 9.2] gives a number of possible design alternatives to cut launch costs. They are: 1. Optimise for minimum cost – Rather than optimising for minimum size and weight as has occurred in the past, optimise for minimum lifecycle cost. This thesis has optimised for minimum lift-off weight but this is only the starting point. A little extra structural strength or little extra internal size for easier maintenance or adjustment on the pad will increase reliability and decrease cost of manufacture but increase size and lift-off weight. Decreased complexity will decrease development costs [Ref 9.13]. 2. Reduce the parts count – The number of parts is a major cost driver. A 1960’s study concluded that a low cost launch vehicle could be developed at 18% the cost of a Saturn V but with greater size and weight. The cost of materials in a Saturn V was only 3% of the overall cost so simplicity of manufacture, reducing labour, could have a major effect on lowering the cost of the vehicle [Ref 9.14]. 3. Increase Simplicity and Margins – Decreasing complexity with increased design margins will allow increased reliability with confidence in operation and will reduce the need for backup systems. Failure modes will then be reduced which will further increase confidence in the launchers performance and reliability. This could be important if launching south out of Beecroft peninsula along the south coast of New South Wales and Victoria. 4. Reduce Instrumentation – Current launch vehicles are heavily instrumented for later diagnosis of faults in case of failure. A simpler more reliable vehicle with greater

128

Chapter 9 System Synthesis

margins would need less instrumentation, less telemetry bandwidth and less staff to store and analyse the data. On-board electronics could use commercial-off-the-shelf (COTS) electronics. The practice of testing and discarding COTS components that fail test produces lower cost than purchasing space rated components. 5. Emphasise Manufacturing – Simplicity of manufacturing and large production runs can decrease the cost of a launch vehicle. Decreasing cost increases demand which further lowers cost bringing the benefits of mass production. An example is the Russian Vostok launch vehicle which uses the RD-107 and RD-108 engine. Each launch vehicle uses 16 RD-107 engines in the strap-on boosters and 4 of the slightly modified RD-108 engines in the central core, a total of 20 identical thrust chambers per vehicle [Ref 9.15]. This enabled the Russians to produce four small engines instead of a single large engine. With over 1500 Vostok boosters launched the modular and high requirement for engines with stage commonality allowed high production rates of components. As a further example of commonality, the Ariane AR44L uses four Viking VC engines in the first stage and one slightly modified as the Viking IVB in the second stage. Chapter six calculations have used five engines in the ALV vehicle in the first stage and one in the second stage. In the ALV, as has been discussed this is intended to allow a certain amount of single engine failure and thrust throttling (step-down) as well as commonality.

9.2.4 Launch Sites A near equatorial launch site would appear to be out of the question due to the presence of the environmentally sensitive Great Barrier Reef and inhabited Pacific Islands to the east of the Australian mainland. Similarly a continental launch is also not possible although a proven reliable launch vehicle may enable range safety requirements to be relaxed in the future. In the meantime, a more southerly launch site on the East coast at Beecroft Peninsula in New South Wales would serve for low inclination launches. The site would need to be shared with its current owners, the Royal Australian Navy, which operates a naval gunnery range head on the site. It may also be possible to operate Polar and Sun-synchronous launches from this site by tracking south along the New South Wales and Victorian coasts.

Figure 9.1: Naval air station, HMAS Albatross at Jervis Bay (Nowra) [Ref 9.15]

129

Chapter 9 System Synthesis

The location of Beecroft Peninsula as a launch site is particularly advantageous because of its location in the middle of the East Coast industrial strip from Melbourne and Geelong in the South to Woollongong, Sydney and Newcastle to the North. The Royal Australian Navy (RAN) port facilities at Jervis Bay (Nowra) are less than 20km distant by air but a little further by road. The RAN air station at Nowra, HMAS Albatross (Figure 9.1), also has major airport facilities and is soon to undergo significant upgrading under the Albatross Stage 3 Redevelopment Project, with the Australian Government about to invest $130 million into upgrading base infrastructure. This upgrade will commence in 2010. In the event that it is not possible to launch Polar or Sun-synchronous missions from Beecroft Peninsula a second launch site would be required. A suitable site can be found at Point Malcolm on the south coast of Western Australia some 173 kilometres East of Esperance. For ISS resupply missions, the Point Malcolm site would definitely be needed as a direct injection ISS mission launched from Beecroft Peninsula would track over the south island of New Zealand. ISS orbit direct injection launches from Point Malcolm would pass to the south of Tasmania and New Zealand and across the empty South Pacific Ocean. 9.2.5 Manufacture While small components such as the electronics, even the engines, can be manufactured anywhere in the country, large components such as the fuel tanks, interstage and engine bay skirts and payload fairings should be manufactured at the launch site to minimise transport costs and problems with oversize objects having to be moved along existing transport routes. The factory for these units can be situated several kilometers from the launch pad and moved to the assembly building by dedicated road or rail and thence to the vertical integration building. Consideration should also be given to the possibility of manufacturing the propellants (Liquid Oxygen/Liquid Hydrogen) at the launch site. Materials: Sellers and Milton [Ref 9.2, Chapter 3] demonstrate that the political environment is one of the biggest cost drivers in a space mission. This is true of materials selection. The decision to use high tech materials may involve purchase from an overseas source. This may involve import and export licenses which may take time to acquire, may be delayed indefinitely and ultimately refused. This is an area which allows political interference with the project even pressure brought to bear by foreign competitors. The choice of toxic materials again allows environmental pressure to be applied. Once again permits may be delayed or refused. The environmental lobby may exert influence for the sake of a principle thus delaying the project while environmental and safety studies are done on even the smallest of components of the project. Explosives or pyrotechnic components such as explosive bolts or linear shaped charges will require licensing and safety analyses. Fortunately DSTO has extensive experience in developing such devices and these devices can be designed and manufactured within Australia. High tech materials such as titanium or composites require expensive processing facilities of which there are often only a limited number available. This again allows political pressure to be used to delay a project, or delay to the schedule due to the machining job being bumped to a lower priority. Instead of high tech materials like titanium, the use of common materials such as aluminium or stainless steel is recommended or a willingness to provide dedicated production facilities. The use of materials that are already in common use in the country that can be diverted to the launch vehicle project is the best option to avoid political pressure and to minimise delays in supply. The participation of production personnel in the design teams from the

130

Chapter 9 System Synthesis very beginning of the project will ensure correct materials choice and production processes are specified to ensure producibility of the components. Risk: The confidence of potential customers is the key to commercial success and a full order book. Engineering determines acceptable risk in terms of achieving the mission. However perceived risk as viewed by the customer is the driving constraint to allowable risk. Reduced documentation and analysis and increased design margins may be acceptable from an engineering point of view but customers may require tighter specifications. Customers may feel the need to require the vehicle to be designed to MIL-SPEC/MIL-STD (or higher: High Rel. or Space Qual.). This should be avoided however as each MIL-STD specification explodes into many more2 at each level of component below that specified [Ref 9.2, page 67]. This adds to the design and production costs. A better procedure is to specify the performance and reliability of each component and to allow the manufacturer to test each component to those specifications. If necessary the essential required parts of the MIL-SPEC can be extracted and inserted into the design specifications. Subsystem testing and burn-in along with full system “all-up” testing will identify those components which will fail, reset or misbehave in actual system operation. Extra non-paying test flights may be required to prove a vehicle designed around lower documentation and analysis and increased design margins. This is particularly true of launch licensing authorities where there is again a possibility of political interference. A totally successful first launch would go a long way to achieving customer confidence. Manpower: A project to develop, manufacture and operate an Australian launch system would require a considerable skilled workforce. If one of the purposes of the project is to stimulate technical spin-offs for the nation as well as provision of a space launch capability a permanent workforce consisting primarily of Australian citizens rather than foreign nationals or contractors is desirable. In this author’s opinion there would be no problem fulfilling the staffing requirements. The author’s experience is that there are many, if not many thousands, of expatriate Australian scientists and engineers working abroad. These persons could be attracted home permanently by a properly constituted and funded project. The large numbers of technical personnel who have left Australian shores to work and live overseas is due to the lack of technical work in this country. This is backed up by author’s experience of working in the UK, Europe and the USA where Australian personnel were encountered working in fields such as space, defence and nuclear physics. Consequently there is a large pool of talented Australian nationals with experience living and working overseas from whom to choose when recruiting for an Australian launch vehicle development and operation project.

2 US DoD has found that each MIL-SPEC refers to another eight MIL-SPECs in a tiering effect increasing exponentially with each level of component and sub component. DoD now requires a permit to use MIL-SPECS [Ref 9.1, page 67]

131

Chapter 9 System Synthesis

9.2.6 Operations The United States Air Force specifies four characteristics that are required of a launch system. They are: capability, reliability, affordability and responsiveness [Ref 9.17]. Carefully designed operational procedures for manufacture, transport, assembly, stacking, payload integration and launch can benefit all of the above desirable characteristics. Capability and reliability have already been covered implicitly or explicitly and are directly functions of design and development. It is self evident that responsiveness is a function not only of design of the vehicle but of the operational procedures of the lifecycle after development from manufacture to launch. As operational costs account for 60% of the lifecycle costs, affordability is also a function of the efficiency of the system operation after development. Interfaces: Spacecraft require support from the launch vehicle during countdown and during the ride into space or during a hold on the pad. These interfaces include electrical (power, command and control), mechanical, fluid and structural elements. There are also non-physical interfaces such as mission analysis, operational plans, safety checks and various requirements of the launch vehicle such as fit checks, weight, static and dynamic balance, coupled load analyses and post-launch reports. Meeting these interface requirements can seriously affect the responsiveness of a launch vehicle system. London [Ref 9.13] gives the example of the Atlas/Centaur vehicle which over the years “sprouted a variety of spacecraft interfaces” increasing the cost and complexity of payload integration by a factor of ten. Atlas/Centaur integration activities started at least 36 months prior to launch [Refs 9.18, 9.19]. Orbital Sciences’ Pegasus launch vehicle has a typical procurement, analysis, integration and test activity cycle of 24-30 months for a baseline payload mission [Ref 9.20]. Arianespace requires a 24 month working baseline for their interface documentation and reviews for both Ariane 4 and 5 [Refs 9.4, 9.5]. An Australian launch vehicle system should aim to equal or better these timelines to provide increased responsiveness and commercial viability. Stage Assembly: It is proposed that the large vehicle components such as fuel tanks should be manufactured at the launch site in order to avoid transport problems. This will be particularly important in the future as the launch vehicle grows in size with increasing payload demands. Other components manufactured elsewhere in Australia would then be brought in by road or air and assembled onto the tanks to form the completed stages. The stages would then be moved by road or dedicated rail to the Vertical Assembly Building (VAB) inside the range. The strategy of constructing the tanks outside the range should be no problem at Beecroft peninsula as a factory can be set up just outside the gunnery range-head. However the relative remoteness from population centres of a site like Point Malcolm would create an expensive overhead for manufacturing the tanks. The duplication of the manufacturing facility is also an undesirable expense in itself. It may be possible to air-lift the stages by heavy lift helicopter, if not by road, from the Beecroft site directly to Point Malcolm or to a seaport, such as the Naval facilities at Jervis Bay, from which the stages could be shipped to Esperance and thence by road to the Point Malcolm launch site.

132

Chapter 9 System Synthesis

Figure 9.2: The Space Launch Complex 6 “Slick Six” for the Delta 4 Heavy launch vehicle at Vandenberg Air Force Base, Ca., USA. The picture shows the mobile assembly shelter on the left (structure with the large American flag) and the mobile service tower on the right. The rocket stands next to the fixed umbilical tower in the centre of the pad complex. Photo: Justin Ray/Spaceflight Now Vehicle Stacking, Checkout & Launch: The United States Apollo program to land a man on the moon during the 1960’s used a process of vertical assembly (stacking) of the Saturn 5 vehicle in the Vertical Assembly Building (VAB) followed by transport of the vehicle to the launch pad atop a giant tracked road “crawler”. The European Ariane 5 uses the same system; however a second system is in use exemplified by the facilities for the Delta-4 launch vehicle at both Cape Canaveral Air Force Station on the US East coast and at Vandenberg Air Force Base on the US west coast. Originally developed in the 1960s for the Titan IIIM vehicle to launch secret military polar orbit spacecraft and later proposed but not used for Shuttle launches, the Space Launch Complex 6 “Slick Six” at Vandenberg (Figure 9.2) has a chequered history of not being used despite a large amount of money being spent on it [Ref 9.21]. That is until it was adopted as the launch complex for the Delta-4 Heavy launch vehicle. The vehicle processing flow adopted at this site was developed to keep the vehicle and payload secret and involves a total enclosure of activities by utilising mobile buildings to totally surround the launch pad. The processing flow used is for major components (stages) to be delivered to a Horizontal Integration Facility (Figure 9.3) where final assembly and checkout is carried out. The stages are then transported horizontally to the launch pad where they enter the Mobile Assembly Shelter (MAS). In the MAS the vehicle is assembled (stacked) horizontally and erected to the vertical by a hydraulically powered erecting cradle (Figure 9.4). The MAS is then withdrawn and the vehicle is left standing on the launch pad with the Mobile Service Tower (MST) and the umbilical tower still servicing the vehicle. At launch the MST is withdrawn leaving the umbilical tower providing essential services until lift-off. Compared with the Apollo program this strategy lowers the cost of transporting the vehicle from VAB to pad and provides an all weather shelter as well as security until launch time.

133

Chapter 9 System Synthesis

Figure 9.3: Kennedy Space Center, Florida. - Two Boeing Delta IV first stages head to the Horizontal Integration Facility (upper right) at Launch Complex 37, Cape Canaveral Air Force Station. Photo: NASA As operations comprise 60% of the lifecycle costs of a launch system it is important to optimise the processing flow at the launch site. The Delta 4 flow is an improvement on the Apollo flow but it is possible that an even more advantageous flow could be developed. Arianespace utilised an enclosed “Coffin” system fitted with air conditioning and various recording devices to measure temperature, humidity and shock loads during delivery of the stages from Europe to Kourou [Ref 9.22]. An attendant also accompanied each stage to assure appropriate treatment of the stage during the trip.

Figure 9.4: The Delta 4 rocket stage is lifted upright in the Mobile Assembly Shelter at Vandenberg's SLC-6 launch pad. Photo: Lloyd Nagle/Boeing

134

Chapter 9 System Synthesis

If completed, tested, certified and coffined stages were to be delivered from the factory [Ref 9.23] directly to the mobile assembly shelter at the launch pad a whole step of the processing flow could be eliminated, bypassing the horizontal integration facility. Suitably “hardened” reliable stages would further enhance the ability to decrease checkout procedures at the launch site. If the MST could close doors to completely enclose the vehicle once the MAS was withdrawn, a 24/7 air conditioned shirt sleeve environment could be provided with 360 degree all around access to the vehicle to ensure protection from weather, ease of access and would assist in minimising time for checkout until launch. Careful design of processing flow from manufacture to launch would serve to enhance system responsiveness.

Figure 9.5: Mobile Service Tower docked with Mobile Assembly Shelter at SLC-6, Vandenberg AFB, Ca., USA. Photo: Ken R. Harman, http://www.spaceistheplace.ca/

9.2.7 Development Facilities: It is the high cost of facilities for development, manufacture and launch (one- time, non-recurring) that presents a major hurdle for the first clean sheet launch vehicle project in any nation. This infrastructure must be funded, normally by governments, with no immediate return. Even after perhaps five years when an engine development is looking as if it will succeed, there is still no start to payback to amortise the facilities until the first successful paying launch. Not all facilities need to be provided at once, however, so expenditure can be spread over a number of years or held over if slow progress is being made. The engine development is the driving element of vehicle development so engine development needs to start early. Once some progress has been made other facilities for the complete vehicle development can be provided. As this thesis is only concerned with the major decisions to be made at the beginning of a project, only those facilities that need to have early planning started are of concern here. The relevant items in order of need are:

135

Chapter 9 System Synthesis

• Instrumented Horizontal Firing Engine Test Bed • Turbopump Test Bed • Vertical Firing Engine (Cluster) Test Stand • Engine Vacuum Start Proving Test Chamber It is recommended that these facilities be constructed to take considerably larger hardware than is immediately required to allow for future size increases. This would avoid such problems as those encountered in developing an upper stage for Blue Streak as mentioned in Chapter 3 [Ref 1.4]. Timeline: The engine development is the driving element in the development schedule and should be started as soon as possible in the program. As it is a generic facility the engine test cell can be constructed before the system is designed. Figure 9.6 shows a possible timeline for the first five years of a system development program. Of the facilities the engine test cell and the turbopump test cell are required at the start of the program, at least by the start of engine development at the 18 month time. The engine vertical test stand, capable of taking a full first stage engine cluster and the engine vacuum start test chamber are required for the third stage engine qualification at the 4 year time frame. Figure 9.7 shows a work breakdown structure for the Phase A concept design phase.

9.3 OPTIMISATION SOFTWARE Given the increased difficulty of obtaining existing orbit and launch trajectory optimisation software due to the current security situation, a number of computer programs were written as part of this thesis work to aid in system planning. A small program was written to calculate the effect of latitude and increasing plane change on payload required to be delivered to geostationary transfer orbit for a given payload into drift orbit. A spreadsheet (STAGEX) was written to calculate the optimal launch vehicle stack for a given total delta-V budget into GTO. Constraints were included to enable lift-off thrust to weight to be specified along with maximum axial loading at stage burnout. A computer program was written in FORTRAN and C++ to duplicate the STAGEX functions utilising a genetic algorithm to carry out the stack optimisation. This program was then extended (TRAJ2DF) to integrate the two dimensional equations of motion during launch with more constraints added. Using the Google Earth GIS, the trajectory and the launch event sequences were then able to be plotted and to present the positions of stage fallout to indicate range safety problems. TRAJ2DF proved to be too slow in computation time on a desktop PC to further develop the optimisation of the launch system. Plans are being considered to speed the operation of TRAJ2DF by implementation on an 1100 CPU cluster supercomputer. However, even at the current stage of development the above suite of software represents an aid to planning for an Australian launch vehicle system.

136

Chapter 9 System Synthesis

137

Chapter 9 System Synthesis

138

Chapter 9 System Synthesis

9.4 RECOMMENDATION SUMMARY FOR AN AUSTRALIAN SPACE LAUNCH SYSTEM The System: • The system should be optimised for minimum lifecycle cost rather than for minimum vehicle lift-off weight • Manufacturing should be located close to launch site to minimise transport for components/personnel • Payload Interfaces should be designed to standards suitable for all payloads • The Payload Integration Schedule should be streamlined for a minimum time line • Stacking and Checkout Procedures should be streamlined for a minimum time line and minimum time on launch pad • COTS Hardware and Software should be utilised where possible. • System should be designed with large margins where possible. • Launch vehicle should be designed for reliability and simplicity • Documentation should be minimised for all lifecycle phases • MILSPEC/STD should not be used unless absolutely necessary. • Relevant parts of MILSPEC should be extracted into specifications The Launch Vehicle: • Nominal mission to Geostationary Transfer Orbit • Nominal Payload capability AR44L • Actual Payload capability AR44L to AR5 (“ten tonne”) t.b.d. • Three stage expendable configuration • No strap-on boosters

• Propellant LOX/LH2 in all stages • Five engine cluster in first stage • Single engine in second stage • Single engine in third stage • Engine commonality between 1st and 2nd stages • Use a nozzle extension skirt on first stage engine to modify for second stage engine use in vacuum • Third stage engine should be capable of multiple restarts • First and second stage engine should be designed from scratch to be restart capable by add-on kit. • First stage engine thrust, 37300 kg.f @ SL • Third stage engine thrust, 6000 kg.f in vacuuo • Steering by Thrust Vector Control • Fairing diameter to take 4.0m OD payload at least • Fairing diameter to take 5.0m OD payload desirable • Gross Lift-off Weight will be approximately 143,459 kg

139

Chapter 9 System Synthesis

• Vehicle design for simplicity, minimum cost and maximum reliability not minimum weight • Spacecraft interfaces should be kept simple • Electronics to be COTS, rated by testing The Launch Site: • Two launch sites will be required • East coast site at Beecroft Peninsula is suitable for low inclination orbits • Polar and sun-synchronous launches may also be possible from Beecroft peninsula • South coast site at Point Malcolm for Polar Sun-synchronous and ISS resupply launches • Large components (e.g. fuel tanks) should be manufactured at the launch site to eliminate transport costs and time • Propellant should also be manufactured at the launch site to eliminate transport costs and time • Two launch pads should be provided at each site for redundancy • Down range tracking and telemetry will be required to be located on shipboard as well as land based

9.5 CONCLUSIONS 1. A commercial launch vehicle system is economically viable and the potential income and employment possibilities exceed those of many existing Australian industries. 2. A conventional three stage stack fuelled by Liquid Hydrogen and Liquid Oxygen is the basic architecture design choice for an Australian launch vehicle system. This is for technology development and environmental reasons. 3. A policy of stressing reliability and simplicity rather than minimum lift-off weight should be adopted and would aid in reducing costs and increasing responsiveness of the system. 4. With past and current Australian industry experience this project is technically achievable with indigenous technical capacity and manpower. However the engine development would be a new experience for the nation. The whole project would require government support perhaps on a loan or guarantee basis but could return income in the operational phase. 5. The Naval gunnery range head at Beecroft peninsula is suitable for a launch site and the higher latitude penalty in terms of increased vehicle lift-off mass for low inclination launches is acceptable when compared with near equatorial sites. Polar and Sun-synchronous launches may also be possible from this site depending on range safety restrictions. 6. A second launch site located at Point Malcolm in WA is a suitable site for polar, Sun- synchronous and ISS resupply launches 7. The Software suite of STAGEX, ROKOPT and TRAJ2DF are useful tools for system planning. TRAJ2DF is too slow to be fully developed on a single desktop PC but can be improved by being implemented on a parallel processor system such as the 1100CPU cluster at the ANU. 8. The penalty for the nation not engaging with a space launch capability is to fall further and further behind in the technological and capability stakes.

140

References

CHAPTER 1 - INTRODUCTION 1.1 Southall, Ivan, Woomera, Angus and Robertson, Sydney, 1962 1.2 Southall, Ivan, Rockets in the Desert, Angus and Robertson, Sydney, 1964 1.3 Morton, Peter, Fire Across the Desert, Woomera and the Anglo-Australian Joint Project, 1946-1980, Australian Government Printing Service for the Department of Defence, Canberra, 1989 1.4 Hill, C.N., A Vertical Empire – The history of the UK Rocket and Space Program, 1950-1971, ISBN 1-86094-267-9, Imperial College Press, London, 2001 1.5 Mach 10 – We made it, Connections, Defence Science and Technology Organisation, ISSN 1328-2050, Number 116, p7, July 2007 1.6 Scramjet continues test success, Connections, Defence Science and Technology Organisation, ISSN 1328-2050, Number 119, pp4-5, October 2007 1.7 Aussat Pty ltd., Aussat: Australia's National Satellite System, 1983, and Wikipedia, Optus Fleet of , http://en.wikipedia.org/wiki/Optus_fleet_of_satellites 1.8 An Australian Study of a Launching Site for Equatorial Orbits, ELDO Project Study 5.4, Adelaide, Department of Supply, December 1965 1.9 Cape York International Spaceport, Part 1 of a feasibility study, Institution of Engineers, Australia, ISBN 0-85825-382-8, February 1987 1.10 MacDonnell Wagner; Hollingsworth Consultants; Crooks Michell Peacock Stewart (Qld) Pty Ltd, Cape York International Spaceport – Infrastructure and Environmental Issues Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-390-9, July 1987 1.11 Touche Ross International, Cape York International Spaceport – Management Processes Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-391-7, July 1987 1.12 Chambers, McNab, Tully & Wilson, Cape York International Spaceport – Legal Issues Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-392-5, July 1987 1.13 Peat, Marwick, Mitchell Services, Cape York International Spaceport – Economic Impact Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-393-3, July 1987 1.14 AUSSAT Pty Ltd, Cape York International Spaceport – Commercial Opportunities Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-394-1, July 1987 1.15 Middleton, B.S., FedSat: Aiming For The Right Orbit?, The Institution of Engineers Australia – Proc. Eleventh National Space Engineering Symposium, University of New South Wales, Sydney, pp. 1-12. 26 February 1997 1.16 Briggs, G.P. et al, Australian Light Launch Vehicle, ALLV - “Capricorn”, Proposal to Queensland Government, Capricorn Launch Services Consortium, September 1987

141

References

1.17 Tuohy, Ian; Franzen, Roger; Raju, John; The Southern Launch Service, The Institution of Engineers, Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 1.18 Goodridge, M.J.; Tuohy, I.R.; Infrastructure for the Southern Launch Service, Institution of Engineers, Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 1.19 Raju, J.A.S.; Bryce, I.R.; Malana, C.; Bondarenko, L.J.; Trade-off Studies in the Concept Definition of the Southern Launch Vehicle, The Institution of Engineers. Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 1.20 Franzen, R.L.; Burbidge, G.T.A.; Arthur, P.; Bollinger, P.; Huxtable, G.; The Southern Launch Vehicle Avionics and the Role of GPS, The Institution of Engineers. Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 1.21 Taurus Launch System Payload User’s Guide, Release 4.0, Orbital Sciences Corporation, March 2006 1.22 Minotaur IV User’s Guide, Release 1.1, Orbital Sciences Corporation, January 2006 1.23 Blair, M.; AUSROC – Australia’s Amateur Rocket Program, The Institution of Engineers. Australia - Seventh National Space Engineering Symposium, Canberra, 21- 26 September 1992 1.24 ASRI, The AUSROC III Program 1.25 Henderson, B., Small Sounding Rocket Program, Australian Space Research Institute Newsletter #11, November 1997

CHAPTER 2 – LITERATURE SURVEY 2.1 Tsiolkovsky, Konstantin E. (1903), "The Exploration of Cosmic Space by Means of Reaction Devices", The Science Review (5), 1903, in Russian 2.2 Goddard, Robert H., “A Method of Reaching Extreme Altitudes” Smithsonian Miscellaneous Collections, Vol 71, No. 2, Washington DC, 1919. 2.3 A Chronology Of Missile And Astronautic Events 1686-1961, Report of the Committee on Science and Astronautics, US House of Representatives, Eighty Seventh Congress First Session, House report No 67, 8 Mar 1967 2.4 Oberth, H., Die Rakete zu den Planetenraumen (The Rocket into Planetary Space) Verlag von R Oldenbourg, Munich, 1923 2.5 Oberth, H., Wege zur Raumschiffahrt (Ways to Spaceflight), Verlag von R. Oldenbourg, Munich, 1929 2.6 Dornberger, W. R., Der Shuss ins Weltall (The Shot into Space), The Viking Press, New York, 1955 2.7 Joubert de la Ferté, Air Chief Marshall Sir Phillip, Rocket, Philosophical Library, New York, 1957 2.8 King, B.; Kutta, T., Impact - The History of Germany’s V-Weapons in World War II, Sarpendon, New York, 1998 2.9 Ordway III, F. I.; Sharpe, M., The Rocket Team, Thomas Y. Crowell, New York, 1979

142

References

2.10 Kraemer, S.K., Rocketdyne: Powering Humans into Space, AIAA Reston Virginia, 2006, ISBN 1-56347-745-8 2.11 McGovern, J., Crossbow and Overcast. New York: W. Morrow, 1964 2.12 Cameron, A.M., Report on Operation BACKFIRE, The War Office, London, Jan 1946 2.13 Wikipedia, “V-2” (html), http://en.wikipedia.org/wiki/V-2 2.14 Wade, Mark. "V-2", Encyclopedia Astronautica. http://www.astronautix.com/lvs/v2.htm 2.15 Dungan, T.D. V-2: A Combat History of the First Ballistic Missile. Westholme Publishing, 2005, ISBN 1-59416-012-0. 2.16 Wade, Mark. "Redstone", Encyclopedia Astronautica, http://www.astronautix.com/lvs/redstone.htm 2.17 Swenson, L.S.; Grimwood, J.M.; Alexander, C.C., This New Ocean, NASA SP-4201, US Government Printing Office, 1966 and, http://www.hq.nasa.gov/office/pao/History/SP-4201/toc.htm 2.18 Grimwood, James M., Project Mercury: A Chronology, NASA SP-4001, US Government Printing Office, 1966 2.19 The Mercury-Redstone Project, Saturn/Apollo Systems Office, George C Marshall Space Flight Center, NASA Huntsville, Alabama, Report TMX-53107, Dec 1964 2.20 Chapman, J.L., Atlas: The Story of a Missile, Victor Gollancz, London, 1960 2.21 Walker, C., Atlas: The Ultimate Weapon by Those Who Built It, Apogee Books, 2005, ISBN 978-1894959186 2.22 Mercury-Atlas, (html), http://en.wikipedia.org/wiki/Mercury-Atlas 2.23 "Mercury Project Summary, Including Results of the Fourth Manned Orbital Flight, NASA SP-45, May 15 and 16, 1963. October 1963. and http://history.nasa.gov/SP-45/cover.htm 2.24 NASA, Project Mercury Home Page, http://www.nasa.gov/mission_pages/mercury/missions/program-toc.html 2.25 Sutton, G.P., History of Liquid Propellant Rocket Engines, AIAA, Reston, Virginia, 2006 2.26 Stumpf, D.K., Titan II: A History of a Cold War Missile Program, University of Arkansas, Fayetteville, 2000 2.27 Wikipedia, “Titan (Rocket Family)” http://en.wikipedia.org/wiki/Titan_%28rocket_family%29 2.28 Wikipedia, “SM-68 Titan”, http://en.wikipedia.org/wiki/LGM-25_Titan 2.29 Wade, Mark, “Titan”, Encyclopedia Astronautica, http://www.astronautix.com/lvs/titan.htm 2.30 Hacker, B.C.; Grimwood, J.M., On the Shoulders of Titans, NASA SP-4203, US Government Printing Office, 1977 and http://history.nasa.gov/SP-4203/toc.htm 2.31 Grimwood, J M., Hacker, B.C., Vorzimmer, P.J., Project Gemini: A Chronology, NASA SP-4002, US Government Printing Office, 1966 and http://history.nasa.gov/SP-4002/cover.htm

143

References

2.32 Bilstein, R.E., Stages to Saturn: A Technological History of the Apollo/Saturn Launch Vehicles, NASA SP-4206, US Government Printing Office, 1980 and http://history.nasa.gov/SP-4206/sp4206.htm 2.33 Apollo Systems Description, Vol II: Saturn Launch Vehicles, Marshall Space Flight Center, Technical Memorandum X-881, Feb 1964 2.34 Saturn V Flight Manual SA-503, MSFC-MAN-503, Marshall Space Flight Center, NASA-TM-X-72151, 1 Nov 1968 2.35 Apollo Technical Data Library, (html) http://www.cs.indiana.edu/sudoc/image_30000061709352/30000061709352/pdf/techdata.htm 2.36 European Space Agency, Building the Soyuz launch facility at Europe’s Spaceport, (html), 18/06/2009, http://www.esa.int/SPECIALS/Launchers_Home/SEMFFUZO0WF_0.html 2.37 Hall, R.D., Shayler, D.J., Soyuz: A Universal Spacecraft, Springer-Verlag, Berlin, 2003, ISBN 1-85233-657-9 2.38 Wade, M., “Soyuz”, Encyclopedia Astronautica, 2008, (html) http://www.astronautix.com/lvs/soyuz.htm 2.39 Baker, D., The Rocket: The History and Development of Rocket and Missile Technology, New Cavendish Books, Sept 1978, ISBN 978-0904568103 2.40 Baker, D., The History of Manned Spaceflight, Crown, New York, 1981. 2.41 Clark, P., The Soviet Manned Space Program, Salamander Books, London, 1988. 2.42 Isakowitz, S.J., Hopkins, J.P., Hopkins, J.B., International Reference Guide to Space Launch Systems Second Edition, AIAA, Washington DC, 1991 (succeeded by 2000 edition). 2.43 Morton, P., Fire Across the Desert, Australian Government Publishing Service, 1989, ISBN 0-644-47500-5 2.44 Parsch, A., GAF Jindivik (html), Directory of U.S. Military Rockets and Missiles, http://www.designation-systems.net/dusrm/app4/jindivik.html , 2004 2.45 Southall, I., Woomera, Angus and Robertson, Sydney, 1962 2.46 Southall, I., Rockets in the Desert, Angus and Robertson, Sydney, 1964 2.47 Sutton, G.P., History of Liquid Propulsion Rocket Engines, AIAA, Reston, Va., 2006 2.48 Hill, C.N., A Vertical Empire: The History of the UK Rocket and Space programme, 1950-1971, Imperial College Press, 2006 2.49 Wade, Mark, “Blue Steel”, Encyclopedia Astronautica, http://www.astronautix.com/lvs/blusteel.htm 2.50 ibid, “Black Knight”, http://www.astronautix.com/lvs/blanight.htm 2.51 ibid, “Blue Streak”, http://www.astronautix.com/stages/blutreak.htm 2.52 ibid, “Europa”, http://www.astronautix.com/lvs/europa.htm 2.53 ibid, “RZ.2”, http://www.astronautix.com/engines/rz2.htm 2.54 Wynn, H., R.A.F. Nuclear Deterrent Forces, H.M. Stationery Office,1994, ISBN 0-11-772833-0

144

References

2.55 Harvey, B., Europe’s Space Program: To Ariane and Beyond, Springer-Praxis, 2003 2.56 An Australian Study of a Launching Site for Equatorial Orbits, ELDO Project Study 5.4, Adelaide, Department of Supply, December 1965 2.57 Cape York International Spaceport, Part 1 of a feasibility study, Institution of Engineers, Australia, ISBN 0-85825-382-8, February 1987 2.58 MacDonnell Wagner; Hollingsworth Consultants; Crooks Michell Peacock Stewart (Qld) Pty Ltd, Cape York International Spaceport – Infrastructure and Environmental Issues Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-390-9, July 1987 2.59 Touche Ross International, Cape York International Spaceport – Management Processes Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-391-7, July 1987 2.60 Chambers, McNab, Tully & Wilson, Cape York International Spaceport – Legal Issues Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-392-5, July 1987 2.61 Peat, Marwick, Mitchell Services, Cape York International Spaceport – Economic Impact Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-393-3, July 1987 2.62 AUSSAT Pty Ltd, Cape York International Spaceport – Commercial Opportunities Scoping Study, Institution of Engineers, Australia, ISBN 0-85825-394-1, July 1987 2.63 Griffy-Brown, C., Knapp, K., Humphreys, J., The Cape York Spaceport: Lessons in Sustainable Strategic Advantage for Developing Commercial Space Ventures, Center for Technology Management, Griffith University, Queensland, Australia, Nov 1998 http://www.gbhap-us.com/fulltext/free/S960040F793.htm 2.64 Google Web Search on Creating a Commercial Launch Industry: What Will it Take for Australia to Succeed?, http://scholar.google.com.au/scholar?q=related:EoP- MGthfPsJ:scholar.google.com/&hl=en&as_sdt=2000 2.65 Briggs, G.P. et al, Australian Light Launch Vehicle, ALLV - “Capricorn”, Proposal to Queensland Government, (unpublished) Capricorn Launch Services Consortium, September 1987 2.66 Wilson, S.A. An Australian Space Launch Vehicle, Fifth National Space Engineering Symposium 1989, Institution of Engineers, Barton, ACT: Australia, 1990. National ISBN 0858254905. 2.67 Tuohy, Ian; Franzen, Roger; Raju, John; The Southern Launch Service, The Institution of Engineers, Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 2.68 Goodridge, M.J.; Tuohy, I.R.; Ground Segment Infrastructure for the Southern Launch Service, Institution of Engineers, Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 2.69 Raju, J.A.S.; Bryce, I.R.; Malana, C.; Bondarenko, L.J.; Trade-off Studies in the Concept Definition of the Southern Launch Vehicle, The Institution of Engineers. Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992

145

References

2.70 Franzen, R.L.; Burbidge, G.T.A.; Arthur, P.; Bollinger, P.; Huxtable, G.; The Southern Launch Vehicle Avionics and the Role of GPS, The Institution of Engineers. Australia - Seventh National Space Engineering Symposium, Canberra, 21-26 September 1992 2.71 Space Activities Act 1998, http://www.austlii.edu.au/au/legis/cth/consol_act/saa1998167/ 2.72 Freeland, S., When Laws Are Not Enough – The Stalled Development Of An Australian Space Launch Industry, University of Western Sydney Law Review, 2004 2.73 Blair, M.; AUSROC – Australia’s Amateur Rocket Program, The Institution of Engineers. Australia - Seventh National Space Engineering Symposium, Canberra, 21- 26 September 1992 2.74 ASRI, The AUSROC III Program (unpublished – available from ASRI), no ISBN 2.75 Henderson, B., Small Sounding Rocket Program, Australian Space Research Institute Newsletter #11, November 1997 2.76 Middleton, B.S., FedSat: Aiming For The Right Orbit?, The Institution of Engineers Australia – Proc. Eleventh National Space Engineering Symposium, University of New South Wales, Sydney, pp. 1-12. 26 February 1997 2.77 Australis-OSCAR-5, http://www.sat-net.com/winorbit/help/satao5.html 2.78 Christmas Island Asia Pacific Launch Facility, http://www.aerospace-technology.com/projects/christmas/ 2.79 Rogers. A.J., Concept Exploration of an Australian Indigenous Space Launch Capability, Air Force Institute of Technology, Wright-Patterson AFB Ohio, School of Engineering Master's thesis, Mar 2001 2.80 Lost in Space? Setting a new direction for Australia's space science and industry sector, Commonwealth of Australia, 2008, ISBN 978-0-642-71996-6 2.81 Gilbert, Jo-Anne, We can lick gravity, but…”: What trajectory for space in Australia?, Space Policy, Vol 25, Issue 3, August 2009, Pages 174-180 2.82 Industry Council to Boost Australian Space Innovation: Government Response to Senate Space Report, Press release by Australian minister for innovation http://minister.innovation.gov.au/Carr/Pages/INDUSTRYCOUNCILTOBOOSTA USTRALIANSPACEINNOVATION.aspx 2.83 Space Industry Council Chair Appointed, Press release by Australian minister for innovation http://minister.innovation.gov.au/Carr/Pages/SPACEINDUSTRYCOUNCILCHAI RAPPOINTED.aspx 2.84 HyShot - Hypersonics News, http://www.uq.edu.au/news/index.html?category=89 2.85 Walker, S., Rodgers. F., Paull, A.,van Wie, D.M., HyCAUSE Flight Test Program, 15th AIAA International Space Planes and Hypersonic Systems and Technologies Conference, 28 April - 1 May 2008, Dayton, Ohio 2.86 Hypersonics Research Speeds Up, Australian Government, Department of Defence, Defence Science and Technology Organisation http://www.dsto.defence.gov.au/attachments/7806%20HyCAUSE%20Fact%20SheetB.pdf 2.87 Michels, J., Przybilski, O., Peenemünde und seine Erben in Ost und West, Bernard & Graefe, Bonn, 1997.

146

References

2.88 Wade, M., Veronique, Encyclopedia Astronautica, http://www.astronautix.com/lvs/vernique.htm 2.89 Ariane 1, 2, 3 and 4 series, Airclaims Space Review, May 2003 2.90 Arianespace, Ariane 4 User’s manual, Issue 2, Feb 1999 2.91 Arianespace Ariane 5 User’s manual, Issue 4, Rev 0, Nov 2004 2.92 Furniss, T., Ariane’s Big Fix, Flight International, 21 Sep 1986, pp. 48-52 2.93 Furniss, T., Ariane 4: The Big Shot, Flight International, 21 May 1988, pp. 30-32 2.94 Dulout, J.P., Utilisation Du Lanceur Ariane (in French), Space Vehicle Orbital Motion, p 473-485, CNES (SEE N81-20108 11-13) 2.95 Breton, J., Ariane 4 Capabilities for Launches into non-Geostationary Orbits, AIAA, 14th International Communication Satellite Systems Conference and Exhibit, Washington, DC, Mar. 22-26, 1992. p.5 2.96 Heydon, A.A., Weinrich, E.H., Ariane 4 and Ariane 5: Performance for the ‘90s and Beyond, AIAA, 14th International Communication Satellite Systems Conference and Exhibit, Washington, DC, Mar. 22-26, 1992. p.8 2.97 Desloire, M., Le Lanceur Ariane: Choix des Matériaux et Problèmes Associés, (in French), Matériaux et Techniques, Oct-Nov 1979, pp. 387-396 2.98 Kemp, R.G., Steering Ariane, Space (ISSN 0267-954X), vol. 3, Nov.-Dec. 1987, p. 40-43. 2.99 Dulout, P.D., Spagnulo, M., The Ariane 4 Launch Vehicle Attitude Control System for in Orbit Payload Separation, AAS 93-277, Advances in the Astronautical Sciences Series (ISSN 0065-3438), p. 407-422 2.100 Waterman, M.D., Guidance Law for the Ariane On-Board Computer, JBIS, Vol.29. pp.402-407, 1976 2.101 Angus, A.M., Gordon, C.S., Brunschvig, A., The Guidance System for Ariane, Journal of Navigation (ISSN 0373-4633), vol. 44, May 1991, p. 159-170 2.102 Cutcher, P.S., Smith, R., Ariane Onboard Flight Control, J.B.I.S., Vol.39, pp.344- 346, 1986 2.103 Souchier, A., Rothmund, C., Leven, P., Hagberg., J., The Development of the Viking Engine, 35th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Los Angeles, CA., 20-24 Jun 1999 2.104 Pouliquen, M., Gill, G.S., Performance Characteristics of the HM7 Rocket Engine for the Ariane Launcher, J.Spacecraft, Vol.16., No.6, pp 367-372, Nov.-Dec. 1979 2.105 Faure, A.P., Development of Cryogenic Upper Stage Tanks (Ariane Launcher – Third Stage Tankage System), ICEC 7; Proceedings of the Seventh International Cryogenic Engineering Conference, London, England, July 4-7, 1978. (A79-31001 12- 31) Guildford, Surrey, England, IPC Science and Technology Press, Ltd., 1978, p. 35-41. 2.106 Guillard, P., de Boisheraud, H., Ariane 5: The H155 Cryogenic Main Stage, ESA Bulletin (ISSN 0376-4265), no. 73, p. 42-53. 2.107 Holsten, H., Development of The Ariane 5 Upper Stage, Acta Astronautica, Vol.29, No.2, pp. 117-120, 1993 2.108 Hudson, G., Private discussion with G.Briggs, 1986

147

References

2.109 Dussollier, G., Teissier, A., Ariane 5 Main Stage Oxygen Tank Pressurization, AIAA 93-1969, AIAA/SAE/ASME/ASEE 29th Joint Propulsion Conference and Exhibit, June 28-30, 1993, Monterey, CA. 2.110 Vernin, H., Pempie, P., Ariane Liquid Booster Trade-Off, AIAA 01-3687, 2.111 Burkhardt, H.; Sippel, M., Herbertz, A., Klevanski, J.,; Kerosene vs Methane: A propellant tradeoff for reusable liquid booster stages, Journal of Spacecraft and Rockets, ,2004, vol. 41,no. 5, pp. 762-769 2.112 JAXA, OHSUMI, http://www.isas.ac.jp/e/enterp/missions/ohsumi.shtml 2.113 JAXA, Lamda-4S Satellite Launch Vehicles, http://www.isas.ac.jp/e/enterp/rockets/vehicles/l-4s/index.shtml 2.114 JAXA, Satellite Launch Vehicles, http://www.isas.ac.jp/e/enterp/rockets/vehicles/index.shtml 2.115 Morita, Y., A World Class Team Develops Cutting-Edge Technology, http://www.jaxa.jp/article/interview/no16/index_e.html 2.116 Kyle, E., N-1, N-2, and H-1: Japan's "Deltas", Space Launch Report, 15/09/2009, http://www.spacelaunchreport.com/thorh11.html 2.117 Wade, M., H-2, Encyclopedia Astronautica, http://www.astronautix.com/lvs/h2.htm 2.118 JAXA, H-IIA Launch Vehicle, http://www.jaxa.jp/projects/rockets/h2a/index_e.html 2.119 JAXA, HTV/H-IIB special Site, http://www.jaxa.jp/countdown/h2bf1/index_e.html 2.120 Wade, M., China, Encyclopedia Astronautica, http://www.astronautix.com/articles/china.htm 2.121 Long March (rocket family), Wikipedia (web resource), http://en.wikipedia.org/wiki/Long_March_%28rocket_family%29 2.122 Gupta, S.C., Suresh, B.N., Sivan, K., Evolution of Indian Launch Vehicle Technologies, Current Science, Vol.93, no.12, 25th December 2007 2.123 Gupta, S.V., Suresh, B.N., Development of Navigation Guidance and Control Technology for Indian Launch Vehicles, in Developments in Fluid mechanics and Space Technology, Indian Academy of Sciences, Bangalore, 1988 2.124 Wade, M., India, Encyclopedia Astronautica, http://www.astronautix.com/country/india.htm 2.125 Wade, M., Scud, Encyclopedia Astronautica, http://www.astronautix.com/lvs/r17.htm 2.126 Scud, Wikipedia (web Resource), http://en.wikipedia.org/wiki/Scud#R-11 2.127 Global Security Organisation, R-11/SS-1B SCUD-A, R-300 9K72 Elbrus/SS-1C SCUD-B, http://www.globalsecurity.org/wmd/world/russia/r-11.htm 2.128 Feickert, A., Iran’s Ballistic Missile Capabilities, CRS Report for Congress, 2004 2.129 Jane's Information Group, SS-1 `Scud' (R-11/8K11, R-11FM (SS-N-1B) and R-17/8K14). 26 April 2001, Retrieved 2008-02-12 http://www.janes.com/security/international_security/news/misc/sws_scud010426.shtml 2.130 Pinkston, D.A., The North Korean Ballistic Missile Program, Strategic Studies Institute, War College, February 2008

148

References

2.131 Barrère, M.; Jaumotte. A.; deVeubeke, B.F.; Rocket Propulsion, Elsevier Publishing Co., Amsterdam, 1960 2.132 Huzel, D.K., Huang, D.H., Design of Liquid Propellant Rocket Engines, 2ed., NASA SP-125, Washington DC, 1967 2.133 Huzel, D.K., Huang, D.H., Modern Engineering for Design of Liquid-Propellant Rocket Engines, AIAA Progress in Astronautics and Aeronautics, AIAA Washington DC, 1992, ISBN 1-56347-013-6 2.134 Crocco, L., Cheng Sin-I, Theory of Combustion Instability in Liquid Propellant Rocket Motors, AGARDograph No.8, Butterworths Scientific Publications, 1956. 2.135 Harrje, D.T., Reardon, F.H., (eds.), Liquid Propellant Rocket Combustion Instability, NASA SP-194, Washington DC, 1974. 2.136 Sutton, G.P., Biblarz, O., Rocket Propulsion Elements, 7ed., Wiley Interscience, NY, 2001, ISBN 0-471-32642-9 2.137 Cornelisse, J.W., Schöyer, H.F.R., Wakker, K.F., Rocket Propulsion and Spaceflight Dynamics, Pitman, 1979, ISBN 0-273-01141-3 2.138 Humble, R.W, Henry, G.N., Larson, W.J., Space Propulsion Analysis and Design, McGraw-Hill, NY, 1995, ISBN 0-07-031329-6 2.139 Kraemer, R.S., Rocketdyne: Powering Humans into Space, AIAA, Reston, Virginia, ISBN 1-56347-754-8 2.140 Dawson, V.P., Bowles, M.F., Taming Liquid Hydrogen: The Centaur Upper Stage Rocket, 1958-2002, NASA SP-2004-4230, Washington DC, 2004, ISBN 0-16073-085-6 2.141 Koelle, H.H. (ed.), Handbook of Astronautical Engineering, 1ed, McGraw-Hill, New York, 1961. 2.142 Fortescue, P., Stark, J. and Swinerd, G. (eds.), Spacecraft Systems Engineering, 3rd Ed, John Wiley & Sons, 2003 2.143 Various Contributors, Space Vehicle Design Criteria, NASA Special Publication Series SP-8001 to 8126, NASA Washington DC, Various publication dates 2.144 Hammond, W.E., Space Transportation: A Systems Approach to Analysis and Design, AIAA Education Series, AIAA, Reston, Va., 1999 2.145 Hammond, W.E., Design Methodologies for Space Transportation Systems, AIAA Education Series, AIAA, Reston, Va., 2001 2.146 Edwards, T., Liquid Fuels and Propellants for Aerospace Propulsion: 1903-2003, Journal of Propulsion and Power, Vol 19, No.6, Nov-Dec 2003, AIAA, Reston, Va. 2.147 Sietzen, F., The Greening of Rocket Propulsion, Aerospace America, July 2005 2.148 Valentian, D., Sippel, M., Grönland, T-A., et al, Green Propellant Options for Launchers, Manned Capsules and Interplanetary Missions, 4th International Conference on Green Propellants for Space Propulsion, 20-21 June 2001, ESA/ESTEC, Noordwijk, the Netherlands. 2.149 Palaszewski, B., Gelled Liquid Hydrogen: A White Paper, Feb 1997, NASA Lewis Research Center, Cleveland OH. 2.150 Gordon, S., McBride, B.J., Computer program for Calculation of Complex Chemical Equilibrium Compositions and Applications, Vol I. Analysis, NASA Reference Publication 1311, Oct 1994

149

References

2.151 Gordon, S., McBride, B.J., Computer program for Calculation of Complex Chemical Equilibrium Compositions and Applications, Vol II. User’s Manual and Program Description, NASA Reference Publication 1311, Jun 1996 2.152 Pratt & Whitney Rocketdyne, Theoretical Performance of Rocket Propellant Combinations (poster) 2.153 Curtis, H.D., Orbital Mechanics for Engineering Students, Elsevier, Oxford, 2006, ISBN-13 978-0-7506-6169-0 2.154 Roy, A.E., Orbital Motion, 4th Ed, Adam Hilger, 2005, ISBN-13 978-0-7503-1015-4 2.155 Bate, R.R., Mueller, D.D., White, J.E., Fundamentals of Astrodynamics, Dover, New York, 1971 2.156 Thomson, W.T., Introduction to Space Dynamics, Dover, New York, 1986 2.157 Larson, W.J., Wertz, J.R., (eds.), Space Mission Analysis and Design, Kluwer, Dordrecht, 1997, ISBN 0-7923-1998-2 2.158 Pocha, J.J., An Introduction to Mission Design for Geostationary Orbits, Reidel, Dordrecht, 1987, ISBN 90-227-2497-2 2.159 Soop, E.M., Handbook of Geostationary Orbits, Reidel, Dordrecht, 1994, ISBN 0- 7923-3054-4 2.160 Wertz, J.R., Larson, W.J, (eds.), Reducing Space Mission Costs, Kluwer, Dordrecht, 2005 2.161 Bevington, P.R., Data Reduction and Error Analysis for the Physical Sciences, McGraw-Hill, 1969 2.162 Press, W.H., Teukolsky, S.A., Vetterling, W.T., Flannery, B.P., Numerical Recipes in Fortran 77: The Art of Scientific Computing, Cambridge, 1999, ISBN 0-521- 43064-X 2.163 Press, W.H., Teukolsky, S.A., Vetterling, W.T., Flannery, B.P., Numerical Recipes in Fortran 90: The Art of Parallel Scientific Computing, Cambridge, 1999, ISBN 0- 521-57439-0 2.164 Nelder, J.A., Mead, R., A Simplex Method for Function Minimization, Computer Journal, 1965, vol 7, pp 308-313 2.165 Zhang, S., Jin, J., Computation of Special Functions, John Wiley and Sons, New York, NY, 1996, ISBN 0-471-11963-6 (includes a disk of the software in FORTRAN) 2.166 Deb, K., Multi-Objective Optimization Using Evolutionary Algorithms, John Wiley & Sons, ISBN 978-0-471-87339-6, 2002 2.167 Deb, K., Optimization for Engineering Design, Prentice-Hall of India, 8ed. July 2005 2.168 Ray, T., Sarker, R., Multiobjective Evolutionary Approach to the Solution of Gas Lift Optimization Problems, Evolutionary Computation, 2006. CEC 2006, IEEE Congress on Evolutionary Computation, pp3182-3188, 2006. ISBN: 0-7803-9487-9 2.169 Deb, K., Pratap, A., Agarwal, S., Meyarivan, T., A Fast and Elitist Multi-objective Genetic Algorithm: NSGAII, IEEE Transaction on Evolutionary Computation, 6(2), pp181-197, 2002 2.170 Ray.T., Liew, K.M., A Swarm Metaphor For Multi-Objective Design Optimization, Eng. Opt., vol. 34, no. 2, pp. 141–153, Mar. 2002.

150

References

2.171 Ross, M., D’Souza, C., Fahroo, F., Ross, J.B., A Fast Approach to Multi-Stage Launch Vehicle Trajectory Optimization, M., AIAA 2003-5639, AIAA Guidance, Navigation, and Control Conference and Exhibit, 11-14 August 2003, Austin, Texas 2.172 Mondoloni, S., A Genetic Algorithm for Determining Optimal Flight Trajectories, AIAA Paper 98-4476,AIAA Guidance, Navigation, and Control Conference and Exhibit, August 1998. 2.173 Wollam, J., Kramer, S., and Campbell, S., Reverse Engineering of Foreign Missiles via Genetic Algorithm, AIAA Paper 2000-0685, 38th Aerospace Sciences Meeting & Exhibit, Reno, NV, January 2000 2.174 Bailey, S.L., Reverse Engineering of a SCUD Missile using a Genetic Algorithm, MS Thesis, Auburn University, May, 14, 2004. 2.175 Metts, J., Hartfield, R., Burkhalter, J., Jenkins, R., Reverse Engineering of Solid Rocket Missiles with a Genetic Algorithm, AIAA Paper 2007-363, 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, 8-11 January 2007 2.176 Riddle, D., Hartfield, R., Burkhalter, J., Jenkins, R., Genetic Algorithm Optimization of Liquid Propellant Missile Systems, AIAA Paper 2007-362, 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, 8-11 January 2007 2.177 Briggs, G.P., Ray, T., Milthorpe, J.F., Evolutionary Algorithm use in Optimisation of a Launch Vehicle Stack Model, AIAA Paper 2007-364, 45th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, 8-11 January 2007 2.178 Geethaikrishnan, C., Multi-Disciplinary Design Optimization Strategy in Multi- Stage Launch Vehicle Conceptual Design, Seminar towards partial fulfillment of the requirement for the award of degree of Doctor of Philosophy (), Department of Aerospace Engineering Indian Institute of Technology, Bombay, August, 2003 2.179 C.Geethaikrishnan, C., Mujumdar, P.M., Sudhakar, K., Adimurthy, V., Genetic Algorithm Guided Gradient Search For Launch Vehicle Trajectory Optimization, Proceedings of the International Conference on Aerospace Science and Technology, 26-28 June 2008, Bangalore, India 2.180 Chartres, J.T.A., Trajectory Design, Optimisation and Guidance for Reusable Launch Vehicles during the Terminal Area Flight Phase, PhD thesis, School of Mechanical Engineering, Faculty of Engineering, Computer and Mathematical Sciences, The University of Adelaide, 2 Feb 2007 2.181 Sobieszczanski-Sobienski, J., Haftka, R.T., Multidisciplinary Aerospace Design Optimization Survey of Recent Developments, Structural Optimization, 14, 1-23 Springer-Verlag, 1997

CHAPTER 3 – SYSTEM CONSIDERATIONS 3.1 Arianespace, Ariane 4 User’s Manual, Issue 2 Rev-0 3.2 Isakowitz, S.J.; Hopkins, J.P. Jr; Hopkins, J.B.; International Reference Guide to Space Launch Systems, 3ed., AIAA, 1999 3.3 Hammock, D.M., Space Vehicles, Principles of Design, Parameters and Criteria for Conceptual Design in Handbook of Astronautical Engineering §22.22, Koelle H. H. ed., McGraw-Hill, New York, 1961

151

References

3.4 Wilson, S.A., Vuletich I.J., Fletcher D.J., et al, Guided Weapon Danger Area & Safety Template Generation – A New Capability, AIAA Atmospheric Flight Mechanics Conference and Exhibit, 18 - 21 August 2008, Honolulu, Hawaii, Paper AIAA 2008-7123 3.5 Space Activities Act 1998, (Australian) Space Licensing and Safety Office Web Site, Australian Space Legislation Page, http://www.innovation.gov.au/General/MEC-SLASO/Pages/AustralianSpaceLegislation.aspx 3.6 ibid, Space Regulations 2001, ibid 3.7 ibid, Space Activities (Approved Scientific or Educational Organisations) Guidelines 2004 3.8 ibid, Flight Safety Code 2nd ed. 3.9 ibid, Maximum Probable Loss Methodology 2nd ed. 3.10 ibid, List of Designated and Protected Assets 3.11 ibid, Administrative Arrangements for the Classification of Assets 3.12 Space Activities Map, (Australian) Space Licensing and Safety Office Web Site, Australian Launch Map Page http://www.innovation.gov.au/General/MEC-SLASO/Pages/AustralianLaunchMap.aspx 3.13 London, J.R., Reducing Launch Costs in Reducing Space Mission Costs, §4.4.5, p130, Wertz, J.R. & Larson, W.J. eds., Space Technology Series, Kluwer and Microcosm, 1996 3.14 Ely, N.; O’Brien, T.P.; Space Logistics and Reliability in Space Mission Analysis and Design (SMAD), 2ed., p694, Larson, W.J. & Wertz, J.R. eds., Space Technology Series, Kluwer and Microcosm, 1992 3.15 ibid, p695 3.16 Roberts, Lt Col T.K., The Need for New Spacelift Vehicles, Space Trace, June 1993

CHAPTER 4 – STRATEGIC PROPELLANT SELECTION 4.1 Ely, N.; O’Brien, T.P.; Space Logistics and Reliability in Space Mission Analysis and Design (SMAD), 2ed., p694, Larson, W.J. & Wertz, J.R. eds., Space Technology Series, Kluwer and Microcosm, 1992 4.2 Sietzen, F. Jr, The Greening of Rocket Propulsion, Aerospace America, July 2005, pp 28-35 4.3 Wade, M., Encyclopaedia Astronautica, 2007, http://www.astronautix.com 4.4 DISR: Department of Industry, Tourism and Resources, 2007, Australian Hydrogen Activity Database, http://www.industry.gov.au/content/itrinternet/cmscontent.cfm? objectID= CFCE5BF1-65BF-4956-B1C29A680ABA6D66 4.5 Edwards, T., Liquid Fuels and Propellants for Aerospace Propulsion: 1903–2003, Journal of Propulsion and Power, Vol. 19, No. 6, November–December 2003, p1089 4.6 Op cit, Wade, http://www.astronautix.com/props/hydazine.htm 4.7 Dunn, Bruce, 1997, Kerosene vs Hydrogen fuelled SSTO rockets, contact: http://www.dunnspace.com/contact.htm 4.8 Dunn, Bruce, 1996, Alternate Propellants for SSTO Launchers, Space Access 96, Phoenix, Arizona, April 25 - 27, 1996, contact: http://www.dunnspace.com/contact.htm

152

References

4.9 Op cit, Wade, http://www.astronautix.com/props/hydazine.htm 4.10 K&Y company, 2002, http://www.hydrazinechina.com/ 4.11 Stumpf, D.K., Titan II – A History of a Cold War Missile Program, Chapter X – Fatal Accidents in the Titan II Program, University of Arkansas, 2000 4.12 Op cit, Wade, http://www.astronautix.com/props/n2o4udmh.htm 4.13 Dornberger, W., V-2, The Inside Story of Hitler’s Secret Weapon, The Viking Press, 1955 4.14 Rosenberg, S.D.; Gage, M.L., Compatibility of Hydrocarbon Fuels with Booster Engine Combustion Chamber Liners, Journal of Propulsion and Power, Dec 1992 4.15 Mc Bride, Bonnie; Gordon, Sanford, Computer Program for Calculation of Complex Chemical Equilibrium Compositions and Applications, NASA-RP1311, Volume I, Analysis, Oct 1994 and Volume II. User manual and Program Description, June 1996 4.16 Isakowitz, S.J.; Hopkins, J.P. Jr; Hopkins, J.B., International Guide to Space Launch Systems, AIAA, 1999 4.17 Sutton, G.P.; Biblarz, O., Rocket Propulsion Elements, 7th edition, Wiley-Interscience, 2001 4.18 Martin, James A., Hydrocarbon Rocket Engines for Earth-to-Orbit Vehicles, Journal of. Spacecraft and Rockets, vol 20, May-June 1983, pp249-256 4.19 Manski, Detlev; Martin, James A., Evaluation of Innovative Rocket Engines for Single-Stage Earth-to-Orbit Vehicles, Journal of Propulsion and Power, Vol 7, Nov- Dec 1991, pp929-937 4.20 Burkhardt, Holger; Sippel, Martin; Herbertz, Armin; Klevanski Josef, Comparative Study of Kerosene and Methane Propellant Engines For Reusable Liquid Booster Stages, 4th International conference on Launcher Technology “Space Launcher Liquid Propulsion”, 3-6 December, 2002 – Liège (Belgium) 4.21 Valentian, D., Sippel, M., Grönland, T., et.al., Green propellants options for launchers, manned capsules and interplanetary missions, Proceedings of the 2nd International Conference on Green Propellants for Space Propulsion (ESA SP-557). 7-8 June 2004, Chia Laguna (Cagliari), Sardinia, Italy. Editor: A. Wilson. Published on CD-ROM. 4.22 Ingersoll, Matthew; Ethanol manager, Manildra, telecon. with G Briggs, 4th Oct 2007 4.23 Atkins, Doreen; Quality manager, CSR Melbourne, telecon. with G Briggs, 4th Oct 2007 4.24 Australian Academy of Technological Sciences and Engineering, Technology in Australia: War-time pharmaceutical chemistry, 21 November 2001 http://www.austehc.unimelb.edu.au/tia/625.html 4.25 Smith, Kate; Account manager, Spectrum Distributors, telecon. with G Briggs, 13th Feb 2008 4.26 DuPont Website, Biobutanol: Fact Sheets, http://www2.dupont.com/Biofuels/en_US/facts/index.html 4.27 Greenbiologics Web Site, Biofuels, 2006, http://www.greenbiologics.com/biofuels.asp 4.28 Wesfarmers Web Site, Wesfarmers Energy: Wesfarmers announces domestic liquefied natural gas plant, 2007, http://www.wesfarmersenergy.com.au/default.aspx?MenuID=29&ContentID=51

153

References

4.29 Kleenheat, New Business–LNG, 2007, http://www.kleenheat.com.au/content/about_us/lng.asp 4.30 Propane, http://en.wikipedia.org/wiki/Propane 4.31 Intergas, R290 Propane, No date, http://www.intergas.co.uk/r290.html 4.32 Burrup Fertilisers Website, World's Largest Ammonia Plant Opens In The Pilbara, 19/4/2006, http://www.bfpl.com.au/index.php?option=com_events&task=article&eid=1&Itemid=7 4.33 Ambrose, Greg; Central Petroleum, telecon with G Briggs, 4th Oct 2007, also Central Petroleum Annual Report(s). 4.34 Huzel, D.K.; Huang, H.H., Design of Liquid Propellant Rocket Engines, NASA SP-125, 1967

CHAPTER 5 – LAUNCH VEHICLE STACK OPTIMISATION BY EVOLUTIONARY ALGORITHM 5.1 Arianespace, Ariane-4 User’s Manual. Issue 2, Rev-0 5.2 White, J. F, ed. Flight Performance Handbook for Powered Flight Operations, John Wiley & Sons Inc., 1963 5.3 Koelle, H. H., ed., Handbook of Astronautical Engineering, §22.23 Multistage Optimisation, McGraw-Hill, 1961 5.4 Ray, Tapabrata; Sarker, Ruhul; Multiobjective Evolutionary Approach to the Solution of Gas Lift Optimization Problem, IEEE Proceedings of the Congress on Evolutionary Computation, Vancouver, July 2006. 5.5 Deb, K., Pratap, A., Agarwal, S. and Meryarivan, T. (2000), A Fast Elitist Non- dominated Sorting Genetic Algorithm for Multi- objective Optimisation: NSGA-II, Proceedings of the Parallel Problem Solving from Nature VI Conference, , France, pp. 849-858 5.6 Deb, K., Pratap, A., Agarwal, S. and Meryarivan, T. (2002), A Fast and Elitist Multi- objective Genetic Algorithm: NSGA-II. IEEE Transaction on Evolutionary Computation, 6(2), pp. 181-197 5.7 Deb, K., Agarwal, S. (1995), Simulated Binary Crossover for Continuous Search Space, Complex Systems, (9) 115-148 5.8 Deb, K.; Goyal, M. (1996), A Combined Genetic Adaptive Search (GeneAS) for Engineering Design, Department of Mechanical Engineering, Indian Institute of Technology 5.9 Press, W. H., Teukolsky, S. A., Vetterling, W. T., Flannery, B. P., Numerical Recipes in Fortran77, second edition, “Chapter 7 - Random Numbers”, Cambridge 1999 5.10 Aliev, R.A., Guirimov, B.G., Bijan Fazlollahi, Aliev, R.R., Evolutionary Algorithm- Based Learning of Fuzzy Neural Networks, Part 2: Recurrent Fuzzy Neural Networks, Fuzzy Sets and Systems 160 (2009) 2553 – 2566 5.11 Nelder, J.A., Mead, R., A Simplex Method for Function Minimization, Computer Journal, 7, 308-313 [1], 1965 5.12 Press, W. H., Teukolsky, S. A., Vetterling, W. T., Flannery, B. P., Numerical Recipes in Fortran77, second edition, “Chapter 10 – Minimization or Maximization of Functions, §10.4”, Cambridge 1999

154

References

CHAPTER 6 – ALTERNATIVE LAUNCH VEHICLE CONCEPTS 6.1 Humble, R.W; Henry, G.N.; Larson, W.J.; Space Propulsion Analysis and Design, Chapter 2-Mission Analysis, McGraw-Hill Space Technology Series, 1995 6.2 Arianespace, Ariane-4 User’s Manual. Issue 2, Rev-0, Feb 1999 6.3 Isakowitz, S.J.; Hopkins, J.P. Jr; Hopkins, J.B.; International Reference Guide to Space Launch Systems, 3rd ed., AIAA, 1999 6.4 Wade, M., Encyclopaedia Astronautica, 2007, http://www.astronautix.com/lvs/ariane.htm 6.5 Arianespace, Ariane-5 User’s Manual. Issue 4, Rev-0, Nov 2004 6.6 Wade, M., Encyclopaedia Astronautica, 2007, http://www.astronautix.com/lvs/saturnv.htm 6.7 Wade, M., Encyclopaedia Astronautica, 2007, http://www.astronautix.com/lvs/ariane5.htm 6.8 Wade, M., Encyclopaedia Astronautica, 2007, http://www.astronautix.com/stages/rla2.htm 6.9 Hammock, D.M., Space Vehicles, Principles of Design, Parameters and Criteria for Conceptual Design in Handbook of Astronautical Engineering §22.22, Koelle H. H. ed., McGraw-Hill, New York, 1961 6.10 Physical Properties (liquids), Department of Earth Observation and Space Systems of the Faculty of Aerospace Engineering University of Delft, http://www.lr.tudelft.nl/live/pagina.jsp?id=58c55f1c-0a06-48cd-ab10- 4ff9426c027d&lang=en 6.11 Desloire, M., Le lanceur ARIANE- Choix des Matériaux et Problèmes Associés, (in French) Matèriaux et Techniques, Oct-Nov 1979 6.12 Saturn V Flight Manual SA-503, MSFC-MAN-503, Marshall Space Flight Center, NASA-TM-X-72151, 1 Nov 1968 6.13 Leitenberger, B, Die Oberstufen H-8, H-10 und ESC-A, http://www.bernd- leitenberger.de/h-10.shtml 6.14 Furniss, T., Ariane 4: the Big Shot, Flight International, 21 May 1988, pp. 30-32 6.15 Roy, A.E., The Foundations of Astrodynamics, Macmillan, New York 1965 6.16 Soop, E.M., Handbook of Geostationary Orbits, Space Technology Library, 1994

CHAPTER 7 – TRAJECTORY OPTIMISATION BY EVOLUTIONARY ALGORITHM 7.1 Thompson, W.T., Introduction to Space Dynamics, Chapter 8 – Performance and Optimization, §8.4 Optimum Program for Propellant Utilization, p252, Dover, 1986 7.2 Humble, R.W; Henry, G.N.; Larson, W.J.; Space Propulsion Analysis and Design, Chapter 2-Mission Analysis, §2.6.3 Steering the Vehicle, p71, McGraw-Hill Space Technology Series, 1995 7.3 White, J.F. (ed), Flight Performance Handbook for Powered Flight Operations, §2.2 Determination of Vehicle Performance, p2-60, John Wiley & Sons, New York, 1963 7.4 Speer, F.A., Flight Evaluation, §28.454 Aerodynamics and Control Systems, specifically fig 28.62, p28-126, in Handbook of Astronautical Engineering, Koelle, H.H. (ed), McGraw-Hill, New York, 1961

155

References

7.5 Huzel, D.K.; Huang, D.H., Design of Liquid Propellant Rocket Engines, NASA SP-125, Chapter 1, First Edition 1967 7.6 Geethaikrishnan, C; Mujumdar, P.M.; Sudhakar, K.; Adimurthy, V, Genetic Algorithm Guided Gradient Search for Launch Vehicle Trajectory Optimization, Proceedings of the International Conference on Aerospace Science and Technology, 26-28 June 2008, Bangalore, India

CHAPTER 8 – LAUNCH SITE AND CORRIDORS 8.1 Space Base proposed for Queensland, ABC On-line news, Friday 26 February 1999, http://www.abc.net.au/science/articles/1999/02/26/19241.htm, 8.2 James, M, Australia in Space: Space Policy and Programs, Parliament of Australia, Parliamentary Library Current Issues Brief 12 1997-98, http://www.aph.gov.au/library/Pubs/CIB/1997-98/98cib12.htm 8.3 Pocha, J.J., An Introduction to Mission Design for Geostationary Satellites, page 11, Space Technology Library, Reidel, Dordrecht, 1987

CHAPTER 9 – SYNTHESIS OF SYSTEM DESIGN CONCEPTS 9.1 Lunar Colony 2018, http://www.kent.ac.uk/physical-sciences/RePh/Lunar%20Colony.pdf 9.2 Wertz. J.R., Larson, W.J., Reducing Space System Costs, Table 4.12, Space Technology Library, Kluwer Academic Publishers, 2005 9.3 Arianespace: Service and Solutions, Launch services, Ariane5 Performance http://www.arianespace.com/launch-services-ariane5/performance.asp, 2008 9.4 Arianespace, Ariane 5 User’s Manual, Issue 4, Rev 0, November 2004 9.5 Arianespace, Ariane 4 User’s Manual, Issue 2, February 1999 9.6 Taverna, Michael A., Europe Urged To Develop Launch Vehicle, Apr 20, 2009 http://www.aviationweek.com/aw/generic/story_channel.jsp?channel=space&id=new s/EADS042009.xml 9.7 Ariane 10 tonnes: AddedLlift Capability for Europe, 27 November 2002 http://www.esa.int/esaCP/ESA26H7708D_index_0.html 9.8 Clark, S, NASA Looking To Solve Medium-Lift Conundrum, Spaceflight Now, August 29, 2009 9.9 Larson, W.J., Wertz. J.R., Space Mission analysis and Design, Space Technology Library, Kluwer Academic Publishers, 1997 9.10 von Karman, T., quoted in Huzel, D.K., Huang D.H., Design of Liquid Propellant Rocket Engines, NASA SP-125, p32, 1967 9.11 Isakowitz, S.J., Hopkins, J.P. Jr., Hopkins, J.B., International Reference Guide to Space Launch Systems, 3ed., AIAA Reston VA., 1999 9.12 NASA. Shuttle Ground Operations Efficiencies/Technology Study Final Report, vol. 6. 1988, NAS 10-11344. Washington, D.C.

156

References

9.13 London, Lt Col John R. III October 1994. Leo on the Cheap - Methods for Achieving Drastic Reductions in Space Launch Costs, Maxwell Air Force Base, AL: Air University Press 9.14 Dergarabedian, P., Nov 14, 1991, Cost Model Considerations for Launch Vehicles, Internal Study, El Segundo, CA: The Aerospace Corporation 9.15 Glushko, V.P., Development of Rocketry and Space Technology in the USSR, Moscow, Novosti Press, 1973 9.16 NAVY: HMAS Albatross http://www.navy.gov.au/HMAS_Albatross#Geographical_and_Demographic_Information 9.17 Roberts, Lt Col T.K., The Need for New Spacelift Vehicles, Space Trace, June 1993 9.18 Holguin, M., Labbee, M., Launch vehicle to Payload Interface Standardization: The Quest for a Low Cost Launch System, AIAA 26th Aerospace Sciences Meeting, Reno, NV, Jan 11-14, 1988 9.19 General Dynamics Space Systems Division, Atlas Centaur Mission Planners Guide, (April 1983, Revised November 1986), Arlington, Virginia and San Diego, California. 9.20 Orbital Sciences Corporation, Pegasus User’s Guide, Release 6.0, Jan 2007 9.21 ‘Slick Six’ launch pad hosts its first Delta 4 Rocket, Atlas Aerospace Industry News, http://www.atlasaerospace.net/eng/newsi-r.htm?id=531, May 27, 2003 9.22 Saunders, P., Tigers in the Jungle in Space: Beyond the Final Frontier, [videorecording], Virtual Worlds, 1997 9.23 Drake, L.R.; Knauf, J.M.; Portanova, P.L., EELV: Evolving toward Affordability, AIAA Aerospace America Online, http://www.aiaa.org/Aerospace/Article.cfm?issuetocid=189&ArchiveIssueID=24

157

Appendix A Sample ROKOPT Data Files

Appendix A ROKOPT Input/Output File

158

Appendix A Sample ROKOPT Data Files

Appendix A1: ROKOPT Input File

HLV: Ariane-44L Equivalent Launch Vehicle 900.0, 4568.0, 200.0, 530.0 ! mfair,mpay,madapt,mebay 29700, 0.09653, 445.1, 0.00 ! mass_s3,sfac_s3,ivac_s3,cf_s3 91600, 0.08997, 293.5, 0.00, 0.00 ! mass_s2,sfac_s2,ivac_s2,cf_s2, minterst_s23 346200, 0.07317, 278.4, 0.00, 0.00 ! mass_s1,sfac_s1,ivac_s1,cf_s1,minterst_s12 ************************************! Optimisation data follows 11945.500, 0.1 ! Mission DV required from the stack, DV tol, m/sec 0, 400000 ! 1st stage mass lower & upper bounds, kg 0, 200000 ! 2nd stage mass lower & upper bounds, kg 0, 60000 ! 3rd stage mass lower & upper bounds, kg ************************************! Evolution parameters follow 50, 50 ! #solutions, #generations 3 40 ! seed, times 0.90, 0.02, 3 ! prob_crosssover: first, step, times 0.05, 0.02, 3 ! prob_mutation: first, step, times 10.0, 5.0, 3 ! dist_index_crossover: first, step, times 10.0, 5.0, 3 ! dist_index_mutation: first, step, times ************************************! Program control switches follow 1,0 ! type out: Run data?, Generation data?: 0=no, 1=yes 1,1 ! print out: Runresults?, History?: 0=no, 1=yes

159

Appendix A Sample ROKOPT Data Files

APPENDIX A2: ROKOPT Output File

HLV: Ariane-44L Equivalent Launch Vehicle 900.0, 4568.0, 200.0, 530.0 ! mfair,mpay,madapt,mebay 29700, 0.09653, 445.1, 0.00 ! mass_s3,sfac_s3,ivac_s3,cf_s3 91600, 0.08997, 293.5, 0.00, 0.00 ! mass_s2,sfac_s2,ivac_s2,cf_s2,minterst_s23 346200, 0.07317, 278.4, 0.00, 0.00 ! mass_s1,sfac_s1,ivac_s1,cf_s1,minterst_s12 ************************************! Optimisation data follows 11945.500, 0.1 ! Mission DV required from the stack,+/- DV tol, m/sec 0, 400000 ! 1st stage mass lower & upper bounds, kg 0, 200000 ! 2nd stage mass lower & upper bounds, kg 0, 60000 ! 3rd stage mass lower & upper bounds, kg ************************************! Evolution parameters follow 50, 50 ! #solutions, #generations 3 40 ! seed, times 0.90, 0.02, 3 ! prob_crosssover: first, step, times 0.05, 0.02, 3 ! prob_mutation: first, step, times 10.0, 5.0, 3 ! dist_index_crossover: first, step, times 10.0, 5.0, 3 ! dist_index_mutation: first, step, times ************************************! Program control switches follow 1,0 ! type out: Run data?, Generation data?: 0=no, 1=yes 1,1 ! print out: Runresults?, History?: 0=no, 1=yes

Run Results File for HLV: Ariane-44L Equivalent Launch Vehicle

Run mass_s1 mass_s2 mass_s3 glow dvdel 1 269172.543 50581.960 38113.305 364065.808 11945.518 1.3 2 179563.303 137018.477 40763.275 363543.055 11945.500 1.0 3 210473.756 102788.814 38665.098 358125.668 11945.731 1.1 4 194599.625 112320.893 44949.523 358068.041 11945.784 1.0 5 205465.554 109820.960 38272.373 359756.887 11948.132 0.9 6 177442.675 163766.967 33445.992 380853.634 11947.888 1.1 7 189572.941 125174.656 40226.870 361172.468 11945.682 1.1 8 262157.497 51821.716 42217.391 362394.604 11945.523 1.1 9 233768.600 74429.779 42282.347 356678.726 11946.054 1.1 10 219940.841 99944.127 34837.475 360920.443 11946.454 1.1 11 240858.463 69913.229 41036.844 358006.536 11950.217 1.1 12 204442.972 103455.253 42878.957 356975.181 11945.616 1.1 13 194460.219 122810.141 38890.075 362358.435 11951.603 1.1 14 268359.113 53390.721 36121.769 364069.603 11946.033 1.1 15 192556.378 115029.548 44602.238 358386.163 11945.526 1.1 16 180208.329 158229.405 33578.489 378214.223 11945.610 1.2 17 228939.421 82436.262 39143.105 356716.788 11945.948 1.1 18 178292.335 126343.585 50780.647 361614.567 11946.104 1.0 19 251159.607 65980.437 36390.618 359728.662 11945.500 1.2 20 198671.944 111878.686 41574.326 358322.956 11946.055 1.1 21 180495.254 136240.460 40566.217 363499.930 11945.753 1.0 22 250665.482 66235.968 36833.114 359932.564 11948.794 1.1 23 185664.947 125405.452 42993.715 360262.114 11945.550 1.1 24 185203.523 127985.556 41831.650 361218.729 11946.479 1.0 25 171884.546 142280.479 43542.760 363905.785 11946.658 1.1 26 236601.781 85381.785 33340.767 361522.333 11947.607 1.2 27 249148.445 62666.038 40739.438 358751.921 11945.548 1.1 28 241473.968 65090.466 46005.449 358767.883 11945.858 1.1 29 209901.377 100528.957 40684.120 357312.453 11947.359 1.1 30 211166.228 100239.417 39710.181 357313.826 11945.500 1.1 31 194110.505 110319.254 47666.672 358294.431 11945.509 1.1 32 215879.382 101199.648 36269.618 359546.647 11945.526 1.1 33 174292.256 135352.763 46324.327 362167.346 11946.727 1.1 34 279156.602 58292.506 28846.503 372493.612 11945.500 1.1 35 193723.035 127935.514 36689.795 364546.344 11946.320 1.2 36 204064.527 101105.524 45690.527 357058.578 11946.296 1.1 37 245036.567 74164.792 34727.319 360126.678 11946.890 1.1 38 284279.203 63852.060 25641.249 379970.513 11946.874 1.1 39 180500.912 133850.925 41973.039 362522.875 11947.079 1.1 40 244017.718 67100.242 40540.617 357856.577 11945.895 1.1 41 180394.408 130106.841 44386.202 361085.451 11946.228 1.2 42 164073.190 142484.789 51969.251 364725.230 11945.504 1.2 43 219715.034 101927.736 34119.124 361959.894 11945.890 1.2 44 223299.429 82348.503 44393.078 356239.010 11945.590 1.2 45 163757.862 145424.051 48981.439 364361.353 11945.500 1.1 46 224428.368 82630.644 43165.102 356422.114 11947.686 1.2 47 233436.789 76461.127 40473.207 356569.123 11945.503 1.0 48 176299.348 131571.412 47441.959 361510.719 11946.317 1.3

160

Appendix A Sample ROKOPT Data Files

49 253623.887 62127.970 37840.526 359790.383 11945.504 1.2 50 226015.439 83400.743 40643.640 356257.821 11945.754 1.5 51 164296.561 144000.336 49774.157 364269.054 11945.560 1.5 52 183066.067 121001.755 50315.427 360581.250 11945.606 1.3 53 289614.826 44421.477 33028.432 373262.735 11947.962 1.1 54 235969.646 76588.550 38651.407 357407.603 11947.474 1.1 55 208282.849 97685.990 44340.223 356507.063 11945.605 1.1 56 255235.248 70989.228 31883.927 364306.403 11949.032 1.1 57 186450.464 117907.566 49207.766 359763.795 11945.798 1.1 58 240486.653 69408.073 41238.824 357331.550 11945.707 1.1 59 207253.755 114677.631 35035.960 363165.347 11945.500 1.1 60 193885.266 113632.153 44479.980 358195.399 11945.601 1.2

Ú 3120 178749.951 145337.193 38048.746 368333.889 11949.478 1.3 3121 205050.460 102158.639 43442.347 356849.446 11945.763 1.3 3122 259053.508 51548.806 46333.841 363134.155 11945.617 1.3 3123 276407.761 44342.982 40034.734 366983.476 11945.817 1.3 3124 161135.787 148822.429 48906.882 365063.097 11945.526 1.2 3125 198911.719 111412.012 42052.802 358574.532 11948.817 1.3 3126 267338.596 58652.846 33418.749 365608.190 11951.860 1.4 3127 202302.804 102984.657 45654.458 357139.919 11945.863 1.2 3128 223155.955 79977.788 48236.038 357567.781 11945.519 1.2 3129 231179.201 76875.633 42160.926 356413.759 11945.798 1.3 3130 231284.417 83768.683 37138.361 358389.460 11950.311 1.3 3131 237287.021 74449.576 39162.104 357096.701 11945.577 1.3 3132 194651.486 115484.133 42355.285 358688.903 11945.898 1.3 3133 271887.253 41768.470 49869.431 369723.154 11945.522 1.3 3134 280083.473 43868.160 38101.993 368251.625 11946.022 1.3 3135 252371.657 61151.217 39766.785 359487.660 11946.360 1.3 3136 241608.539 69086.785 40680.065 357573.389 11946.451 1.3 3137 246489.470 73182.949 34545.521 360415.940 11946.996 1.3 3138 266018.021 54514.556 36483.287 363213.864 11945.515 1.3 3139 175692.476 131264.032 48380.520 361535.028 11945.594 1.4 3140 228331.193 78168.834 43814.774 356512.801 11946.432 1.3 3141 159099.747 145556.251 57549.241 368403.239 11945.510 1.3 3142 237424.942 77500.148 37059.816 358182.907 11947.672 1.3 3143 202904.199 101021.535 47218.594 357342.328 11945.685 1.3 3144 231825.204 80427.590 38530.786 356981.580 11945.778 1.2 3145 238366.702 67466.226 46336.332 358367.261 11945.668 1.3 3146 186599.961 122461.873 44426.920 359686.754 11946.309 1.3 3147 199947.592 106774.434 44467.981 357388.007 11945.973 1.3 3148 253735.654 64955.968 35631.879 360521.501 11945.539 1.4 3149 199393.157 115869.066 38897.296 360357.519 11947.649 1.3 3150 231091.248 76510.538 42753.293 356553.080 11946.439 1.3 3151 237559.405 68950.267 44972.022 357679.694 11945.532 1.4 3152 202535.346 110377.128 39631.880 358742.354 11945.826 1.3 3153 269120.206 54568.823 34873.325 364760.354 11946.784 1.3 3154 246075.451 64752.113 41177.015 358202.579 11945.561 1.3 3155 236673.996 77496.079 37322.148 357690.223 11945.589 1.3 3156 230220.233 73356.234 48616.052 358390.519 11945.552 1.3 3157 229476.293 82107.382 39105.722 356887.397 11946.893 1.3 3158 240260.645 69781.963 41182.771 357423.378 11946.603 1.4 3159 225304.944 80131.391 44843.252 356477.586 11945.903 1.2 3160 237798.348 70390.194 42746.504 357133.047 11945.614 1.3 3161 226201.914 82144.678 41602.850 356147.443 11945.728 1.3 3162 163224.951 146210.970 48868.001 364501.922 11945.500 1.4 3163 239243.967 77595.711 35892.648 358930.326 11947.286 1.4 3164 254669.842 84171.935 27485.497 372525.274 11945.500 1.1 3165 188494.903 118638.179 45671.804 359002.886 11945.616 1.3 3166 258648.941 55104.393 41147.896 361099.231 11945.501 1.3 3167 237738.661 74834.980 38516.106 357287.747 11945.501 1.3 3168 231915.084 84386.368 35992.194 358491.646 11945.823 1.3 3169 302705.154 29954.593 41212.987 380070.734 11945.862 1.3 3170 235399.581 79806.081 36650.911 358054.572 11945.888 1.3 3171 258243.189 51790.291 47013.839 363245.319 11945.529 1.3 3172 254477.889 70686.198 32123.166 363485.253 11946.025 1.3 3173 235672.095 70933.047 44602.149 357405.290 11946.285 1.3 3174 198223.039 107143.842 46141.080 357705.961 11946.351 1.3 3175 236502.956 76314.018 38260.362 357275.337 11945.502 1.2 3176 242497.440 73973.214 36345.417 359014.070 11948.213 1.4 3177 221647.639 90831.198 38482.079 357158.916 11945.978 1.3 3178 239026.423 74354.326 37990.798 357569.547 11945.523 1.3

161

Appendix A Sample ROKOPT Data Files

3179 220266.636 89388.838 40500.030 356353.504 11945.814 1.3 3180 252481.876 56485.960 46007.794 361173.630 11945.952 1.3 3181 211706.244 99488.376 39857.989 357250.609 11945.947 1.2 3182 190198.472 120241.322 42721.024 359358.817 11945.500 1.3 3183 191218.838 116586.945 44718.839 358722.622 11946.201 1.3 3184 217525.607 95992.819 38002.753 357719.179 11945.548 1.2 3185 247510.725 62186.026 42989.694 358884.445 11946.066 1.3 3186 212277.370 99212.479 39574.702 357262.551 11945.513 1.2 3187 243310.032 67991.888 40213.801 357713.721 11945.657 1.2 3188 245104.492 62363.655 45675.595 359341.742 11946.068 1.3 3189 217719.402 87805.965 44434.380 356157.748 11945.591 1.4 3190 177730.377 144999.808 38426.359 367354.544 11946.060 1.3 3191 163515.864 146648.180 48169.445 364531.489 11945.835 1.3 3192 167143.037 164803.550 37043.495 375188.082 11947.593 1.3 3193 204280.801 105304.379 41610.363 357393.543 11945.884 1.3 3194 251968.612 63781.606 37576.696 359524.914 11945.632 1.3 3195 235318.779 68909.807 48649.992 359076.577 11945.593 1.3 3196 237110.385 68822.593 46124.956 358255.934 11946.924 1.3 3197 232045.475 79369.700 39134.245 356747.421 11945.517 1.3 3198 282220.492 44117.435 36619.817 369155.745 11946.258 1.3 3199 210974.594 119054.653 32231.506 368458.753 11946.188 1.4 3200 194940.048 111144.603 45829.928 358112.579 11946.400 1.3 3201 208195.615 112862.760 35285.718 362542.092 11945.768 1.3 3202 221191.458 86440.734 42440.348 356270.539 11947.162 1.4 3203 207748.913 99937.583 42756.503 356640.999 11945.582 1.3 3204 241998.350 71819.342 37957.112 357972.804 11946.180 1.2 3205 238211.784 76354.104 37259.830 358023.718 11946.969 1.3 3206 165054.734 148592.483 45068.527 364913.744 11945.506 1.3 3207 191347.245 114130.739 46878.521 358554.505 11945.564 1.4 3208 204624.101 102118.607 44161.249 357101.958 11947.286 1.3 3209 126871.278 193822.047 49377.091 376268.415 11945.632 1.3 3210 233803.609 75378.856 41256.014 356636.479 11946.051 1.3 3211 178762.988 138921.025 40434.497 364316.510 11946.336 1.3 3212 245461.333 67006.696 39628.841 358294.870 11947.258 1.4 3213 243147.911 66016.343 42527.756 357890.010 11945.567 1.3 3214 190508.855 118310.638 43878.239 358895.732 11945.503 1.3 3215 220800.859 84283.318 45198.322 356480.499 11946.501 1.3 3216 228980.591 83350.513 38612.985 357142.089 11947.373 1.3 3217 157407.044 166691.875 41103.098 371400.017 11945.543 1.3 3218 240899.551 76828.420 35277.666 359203.637 11945.556 1.3 3219 249671.032 67721.637 36048.083 359638.753 11945.508 1.2 3220 228386.171 91856.470 34170.178 360610.819 11945.606 1.3 3221 214541.442 92486.280 43067.660 356293.381 11946.485 1.3 3222 250568.793 61221.519 41126.227 359114.539 11945.715 1.3 3223 194986.058 110938.745 45897.310 358020.114 11945.894 1.3 3224 244571.962 67161.684 40000.585 357932.231 11945.784 1.3 3225 221689.183 88309.800 40176.713 356373.696 11945.629 1.3 3226 212279.800 89904.187 49506.914 357888.901 11945.723 1.4 3227 220718.132 86291.966 43078.247 356286.344 11947.233 1.3 3228 214388.994 90688.249 45149.300 356424.542 11946.275 1.3 3229 169822.396 134527.212 53848.175 364395.783 11946.171 1.3 3230 240790.266 66901.879 43953.938 357844.083 11945.518 1.3 3231 221405.745 86155.096 42255.458 356014.299 11945.503 1.3 3232 220963.485 87380.499 41632.914 356174.898 11946.185 1.3 3233 243967.551 61163.454 50521.367 361850.371 11945.882 1.3 3234 239200.667 67584.870 45020.358 358003.894 11945.965 1.3 3235 182060.055 123536.807 48667.480 360462.341 11946.346 1.3 3236 184818.898 123332.706 45368.109 359717.713 11945.502 1.3 3237 228888.246 82880.226 38946.419 356912.891 11946.759 1.3 3238 235060.127 78972.615 37379.401 357610.143 11945.764 1.3 3239 250804.768 61545.242 40527.055 359075.065 11945.510 1.5 3240 198812.478 107492.780 44929.019 357432.278 11945.501 1.3

3240 Optimisation Runs Completed Total Elapsed Time=2753.3 seconds

Maximum GLOW vehicle: 205023.441 156568.580 27393.705 395183.727 Minimum GLOW vehicle: 220763.908 86454.965 42589.372 356006.245 Mean GLOW vehicle : 220557.306 92627.066 41184.723 360567.094 11972.540 Median GLOW vehicle : 221683.591 88795.552 41659.355 358336.498 11959.738

162

Appendix B Sample TRAJ2DF Data Files

Appendix B1 TRAJ2DF Input File

163

Appendix B1 Sample TRAJ2DF Data Files

Appendix B1: TRAJ2DF Input File

Hypothetical Launch Vehicle ********************************! PAYLOAD DATA FOLLOWS 900.0, 4768.0, 200.0, 400.0 ! FairingMass,kg;SCMass,kg;SCAdapterMass,kg;VeEBayMass,kg ********************************! THIRD STAGE ENGINE DATA FOLLOWS HM7B ! Engine Name LOX/LH2 ! Third stage propellant 3.5, 1.217, 12.0, 3350, 0.66 ! S3ChamberPressure,MPa;S3Gamma,S3PGMW,S3CTemp,K; S3Lstar,m 1.6, 62.5, 14.7 ! S3ContractionRatio; S3ExpansionRatio; S3Pamb,kPa 0.975, 1.01 ! S3etaVstar,S3etaF 20.00, 75.0 ! S3THA,deg; S3BellP,% 3, 15000, 1 ! S3TMode, S3ModeParam, S3NumEngines 445.1 ! S3EstIvac,sec 0.0 ! S3 Turbopump Consumption % of Chamber Propellant Flow ********************************! THIRD STAGE DATA FOLLOWS 13.0, 30721.0 ! S3StructureFactor,%; S3Mass,kg 0.0 ! S3ResDV, m/sec ********************************! SECOND STAGE ENGINE DATA FOLLOWS Viking IVB ! Engine Name N2O4/UH25 ! Second stage propellant 5.85, 1.321, 22.67, 3572, 1.15 ! S2ChamberPressure,MPa;S2Gamma;S2PGMW;S2CTemp,K;S2Lstar,m 1.6, 30.8, 0.0 ! S2ContractionRatio,S2ExpansionRatio,S2Pamb 0.975, 0.983 ! S2etaVstar,S2etaF 20.0, 80.0 ! S2THA,S2BellP 3, 147600, 1 ! S2TMode, S2ModeParam, S2NumEngines 293.5 ! S2EstIvac 0.0 ! S2 Turbopump Consumption of Total Propellant Flow, % ********************************! SECOND STAGE DATA FOLLOWS 9.00, 91010.7 ! S2StructureFactor,%; S2Mass,kg 100.0 ! S2ResDV, m/sec ********************************! FIRST STAGE ENGINE DATA FOLLOWS Viking VC ! Engine Name N2O4/UH25 ! First stage propellant 5.85, 1.335, 22.67, 3572, 1.15 ! S1ChamberPressure,MPa;S1Gamma;S1PGMW;S1CTemp,K;S1Lstar,m 1.6, 12.2, 101.325 ! S1ContractionRatio,S1ExpansionRatio,S1Pamb 0.975, 0.983 ! S1etaVstar,S1etaF 20.0, 80.0 ! S1THA,S1BellP 2, 125000, 5 ! S1TMode, S1ModeParam, S1NumEngines 278.4 ! S1EstIvac 0.0 ! S1 Turbopump Consumption % of Chamber Propellant Flow ********************************! FIRST STAGE DATA FOLLOWS 7.00, 344480.3 ! S1StructureFactor,%; S1Mass,kg 0.0 ! S1ResDV ********************************! GUIDANCE PARAMETERS FOLLOW 5.0 ! Vehicle drag diameter, m; 7.2 10.5 0.1 ! Vertical rise Time,sec; Pitch over angle, deg, MaxAA,deg 5.0, 35.4 ! S1/S2, S2/S3 Interstage coast times, sec 2 48.1 18.6 ! S2 Pitch Program,(1=Gravity turn,2=LTS);S2LTSCoA,S2LTSCoB 2 7.8 0.2 ! S3 Pitch Program,(1=Gravity turn,2=LTS);S3LTSCoA,S3LTSCoB 2 ! Fairing Jettison Criteria, 1=Aerothermal flux, 2=Altitude 110000.0 ! Fairing Jettison Criteria Value ********************************! MISSION - TRAJECTORY/ORBIT REQUIREMENTS FOLLOW 0 ! Mission 200.0 42164.5 ! Perigee Height,km; Apogee Radius,km 100.0 ! Perigee Height=Orbit Attained (km) ********************************! LAUNCH SITE FOLLOWS CSG, Kourou ! Launch Site name ELA-2 ! Pad 7.0, 0.0 ! Latitude, Longitude ********************************! PROGRAM CONTROL PARAMETERS FOLLOW: 1=yes, 0=no 0 ! Type back input data? (pcx1) 1 ! Write back input data? (pcx2) 1 ! Write out engine designs? (pcx3) 1 ! Write out vehicle Stack? (pcx4) 1 ! Write out guidance parameters (pcx5) 1 ! Fly Trajectory=1, Design Vehicle Only=0 (pcx6) 0 ! Include walking impact point? (pcx7) 1 ! Include Coast to impact? (pcx8) 1 ! Write out trajectory=1, GroundZero only=0? (pcx9) 1 ! Write out S1 vertical rise data? (pcx10) 1 ! Write out S1 pitch over data? (pcx11) 1 ! Write out S1 gravity turn data? (pcx12) 1 ! Write out S1 coast to Impact data? (pcx13)

164

Appendix B1 Sample TRAJ2DF Data Files

1 ! Write out S1/2 Interstage Coast? (pcx14) 1 ! Write out S2 Powered Flight Phase? (pcx15) 1 ! Write out S2/3 Interstage Coast? (pcx16) 1 ! Write out S2 coast to Impact data? (pcx17) 1 ! Write out Fairing coast to Impact data? (pcx18) 1 ! Write out S3 Powered Flight Phase? (pcx19) 1 ! Print Master Switch:(pcx20:0=No printing,1=Print as Reqd) 1 ! Write out S3 Coast to Perigee data? (pcx21) ********************************! INTEGRATION TIME STEPS AND OUTPUT STEPS FOLLOW 0.01, 100 ! S1VerticalRiseTime integration Step (sec), Print Step 0.01, 100 ! S1PitchOver integration step (sec), Print Step 0.01, 100 ! S1Gravity turn integration step (sec), Print Step 0.01, 100 ! S1 Coast to Impact integration step (sec), Print Step 0.01, 100 ! S1/2 Interstage Coast integration step (sec), Print Step 0.001, 1000 ! S2 Powered Flight integration step (sec), Print Step 0.001, 1000 ! S2 Coast to Impact integration step (sec), Print Step 0.001, 1000 ! S2/3 Interstage Coast integration step (sec), Print Step 0.001, 1000 ! S3 Powered Flight integration step (sec), Print Step ********************************! OPTIMISATION DATA FOLLOWS 0 ! Carry out Optimisation (0=No, 1=Yes) ********************************! OPTIMISATION VARIABLES 342000 349000 ! First stage Mass: Min, Max (kg) 89000 94000 ! Second Stage Mass: Min, Max (kg) 28000 31000 ! Third Stage Mass: Min, Max (kg) 10.0 14.0 ! First Stage Engine Expansion Ratio: Min, Max 7.0 10.0 ! First stage vertical rise time: Min, Max (sec) 10.0 12.0 ! First stage Pitch over: Min, Max (deg) 30.0 50.0 ! Second Stage Initial Pitch Angle: Min, Max (deg) 10.0 30.0 ! Second Stage Final Pitch Angle: Min, Max (deg) 5.0 10.0 ! Third Stage Initial Pitch Angle: Min, Max (deg) 0.0 5.0 ! Third Stage Final Pitch Angle: Min, Max (deg) 30.0 240.0 ! S2/3 Interstage Coast time, Min, Max (Sec) ********************************! OPTIMISATION CONSTRAINTS 200.0 4.0 ! Perigee Height: Hp, HpTol (km) 42164.5 10.0 ! Apogee Radius: Ra, RaTol (km) ********************************! OPTIMISATION PARAMETERS 150 100 ! #solutions, #generations 59 19 ! integer seed, times 0.90, 0.02, 3 ! prob_crosssover: first, step, times 0.05, 0.02, 3 ! prob_mutation: first, step, times 10.0, 5.0, 3 ! dist_index_crossover: first, step, times 10.0, 5.0, 3 ! dist_index_mutation: first, step, times ********************************! Program control switches follow 1,1 ! type out: Run data?, Generation data?: 0=no, 1=yes 1,1 ! print out: Runresults?, History?: 0=no, 1=yes

165

Appendix B2 Sample TRAJ2DF Data Files

Appendix B2: TRAJ2DF Output File – Flight Mode

Traj2dF - Input Data

Case Study: Hypothetical Launch Vehicle

PAYLOAD AND ACCOMODATION Fairing Mass 900.0 kg Spacecraft Mass 4768.0 kg Adapter Mass 200.0 kg Equipment Bay Mass 400.0 kg

THIRD STAGE ENGINE DATA Engine HM7B Propellant LOX/LH2 Chamber Pressure 3.5 MPa Gamma 1.217 Gas Molecular Weight 12.00 Combustion Temperature 3350°K Lstar 0.660 m Contraction Ratio 1.6 Expansion Ratio 62.5 etaVstar 0.975 etaF 1.010 Throat Half Angle 20.000 deg Nozzle Bell Percent 75.000 Thrust Mode 3 (Specify Vacuum Thrust) Vacuum Thrust 15000 kgf Number of Engines 1 Estimated Ivac 445.1 sec Turbopump Consumption 0.0%

THIRD STAGE DATA Structure Factor 13.0% Stage Mass 30721 kg Reserve Delta-V 0.0 m/sec

SECOND STAGE ENGINE DATA Engine Viking IVB Propellant N2O4/UH25 Chamber Pressure 5.8 MPa Gamma 1.321 Gas Molecular Weight 22.67 Combustion Temperature 3572°K Lstar 1.150 m Contraction Ratio 1.6 Expansion Ratio 30.8 etaVstar 0.975 etaF 0.983 Throat Half Angle 20.000 deg Nozzle Bell Percent 80.000 Thrust Mode 3 (Specify Vacuum Thrust) Vacuum Thrust 147600 kgf Number of Engines 1 Estimated Ivac 293.5 sec Turbopump Consumption 0.0%

166

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Input Data

Case Study: Hypothetical Launch Vehicle

SECOND STAGE DATA Structure Factor 9.0% Stage Mass 91010 kg Reserve Delta-V 100.0 m/sec

FIRST STAGE ENGINE DATA Engine Viking VC Propellant N2O4/UH25 Chamber Pressure 5.8 MPa Gamma 1.335 Gas Molecular Weight 22.67 Combustion Temperature 3350°K Lstar 1.150 m Contraction Ratio 1.6 Expansion Ratio 12.2 etaVstar 0.975 etaF 0.983 Throat Half Angle 20.000 deg Nozzle Bell Percent 80.000% Thrust Mode 2 (Specify Sea Level Thrust) Sea Level Thrust 125000 kgf Number of Engines 5 Estimated Ivac 278.4 sec Turbopump Consumption 0.0%

FIRST STAGE DATA Structure Factor 7.0% Stage Mass 344480 kg Reserve Delta-V 0.0 m/sec

167

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Guidance and Mission Data

Case Study: Hypothetical Launch Vehicle

GUIDANCE PARAMETERS First Stage Vertical Rise Time 7.2 sec First Stage Pitch Over Angle 10.5 deg First Stage Max Angle of Attack 0.1 deg S1/S2 Interstage Coast Time 5.0 sec S2 Pitch Program 2=Linear Tangent Steering S2 Initial Pitch Angle 48.1 deg S2 Final Pitch Angle 18.6 deg S2/S3 Interstage Coast Time 35.4 sec S3 Pitch Program 2=Linear Tangent Steering S3 Initial Pitch Angle 7.8 deg S3 Final Pitch Angle 0.2 deg Fairing Jettison Criteria 2=Altitude Fairing Jettison Value 110000.0 metres

MISSION PARAMETERS SPECIFIED Mission Mode 0 Perigee Height 200.0 km Apogee Radius 42164.5 km

LAUNCH SITE DATA Launch Site name CSG Pad ELA-2 Latitude 7.000 Longitude 0.000

168

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Program Control Switches

Case Study: Hypothetical Launch Vehicle

PROGRAM CONTROL SWITCHES (1=Yes, 0=No) Type back input data (pcx1)? 0 Write input data (pcx2)? 1 Write Engine Designs (pcx3)? 1 Write Vehicle Stack (pcx4)? 1 Write Guidance Parameters (pcx5)? 1 Fly?, or only Design Vehicle (pcx6)? 1 Calculate Walking Impact Point (pcx7)? 0 Calculate Coast to Impact Data (pcx8)? 1 Write Trajectory (pcx9)? 1 Write S1 Vertical Rise (pcx10)? 1 Write S1 Pitch Over (pcx11)? 1 Write S1 Gravity Turn (pcx12)? 1 Write S1 Coast to Impact (pcx13)? 1 Write S1/2 Interstage Coast (pcx14)? 1 Write S2 Powered Flight (pcx15)? 1 Write S2/3 Interstage Coast (pcx16)? 1 Write S2 Coast to Impact (pcx17)? 1 Write Fairing Coast to Impact (pcx18)? 1 Write S3 Powered Flight Phase (pcx19)? 1 Print Master Switch (pcx20)? 1 Write S3 Coast to Perigee (pcx21)? 1

INTEGRATION TIME STEPS S1VerticalRise Integration Step 0.010 sec S1PitchOver Integration Step 0.010 sec S1Gravity Turn Integration Step 0.010 sec S1 Coast to Impact Integration step 0.010 sec S1/2 Interstage Coast Integration Step 0.010 sec S2 Powered Flight Integration Step 0.001 sec S2 Coast to Impact Integration step 0.001 sec S2/3 Interstage Coast Integration Step 0.001 sec S3 Powered Flight Integration Step 0.001 sec

PRINT STEPS (Print every N time steps) S1VerticalRise Print Steps 100 S1PitchOver Print Steps 100 S1Gravity Turn Print Steps 100 S1 Coast to Impact Print Steps 100 S1/2 Interstage Coast Print Steps 100 S2 Powered Flight Print Steps 1000 S2 Coast to Impact Print Steps 1000 S2/3 Interstage Coast Print Steps 1000 S3 Powered Flight Print Steps 1000

169

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Engine Design for Performance

Case Study: Hypothetical Launch Vehicle

THIRD STAGE ENGINE DESIGN Engine HM7B Propellant LOX/LH2 Chamber Pressure 3.5 MPa Gamma 1.217 Gas Molecular Weight 12.00 Combustion Temperature 3350°K Lstar 0.660 m Contraction Ratio 1.6 Expansion Ratio 62.5 etaVstar 0.975 etaF 1.010 Throat Half Angle 20.000 deg Nozzle Bell Percent 75.000 Ambient Operating Pressure 14.700 kPa Thrust Mode 3 (Vacuum Thrust Specified) Chamber Propellant Flow 33.7 kg/sec Sea Level Thrust 12945 kgf Vacuum Thrust 15000 kgf Atmospheric Thrust Loss 2055 kgf Atmospheric Thrust Loss 13.7% Number of Engines on Stage 1 Estimated Ivac 445.1 sec Nozzle Stagnation Mach No 0.40 Injector Total Pressure 3.8 MPa Nozzle Inlet Total Pressure 3.5 MPa Nozzle Inlet Static Pressure 3.2 MPa Throat Static Pressure 2.0 MPa Nozzle Exit Static Pressure 3.9 kPa Chamber Temperature 3350.0°K Nozzle Inlet Temperature 3291.7°K Throat Temperature 3022.1°K Nozzle Exit Temperature 994.7°K Theoretical Characteristic Velocity 2337.4 m/sec Actual Characteristic Velocity 2278.9 m/sec Theoretical Vacuum Thrust Coefficient 1.899 Actual Vacuum Thrust Coefficient 1.916 Theoretical SL Thrust Coefficient 1.637 Actual SL Thrust Coefficient 1.653 Vacuum Exhaust Velocity 4439.5 m/sec Sea Level Exhaust Velocity 3825.9 m/sec Chamber Theoretical Vac Specific Impulse 452.7 sec Chamber Design Vacuum Specific Impulse 445.2 sec Chamber Theoretical SL Specific Impulse 390.1 sec Chamber Design SL Specific Impulse 384.2 sec Turbopump Propellant Usage 0.0% TurboPump Propellant Flow 0.0 kg/sec Engine Total Propellant Flow 33.7 kg/sec Engine Design SL Specific Impulse 384.2 sec Engine Design Vacuum Specific Impulse 445.2 sec

170

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Engine Design for Performance

Case Study: Hypothetical Launch Vehicle

THIRD STAGE ENGINE SIZING Chamber Length 0.346 m Chamber Diameter 0.211 m Length Converging Nozzle 0.083 m Chamber Total Length 0.429 m Throat Diameter 0.167 m Nozzle Bell Length 1.618 m Nozzle Exit Diameter 1.321 m Engine Total Length 2.047 m

SECOND STAGE ENGINE DESIGN Engine Viking IVB Propellant N2O4/UH25 Chamber Pressure 5.8 MPa Gamma 1.321 Gas Molecular Weight 22.67 Combustion Temperature 3572°K Lstar 1.150 m Contraction Ratio 1.6 Expansion Ratio 30.8 etaVstar 0.975 etaF 0.983 Throat Half Angle 20.000 deg Nozzle Bell Percent 80.000 Ambient Operating Pressure 0.000 kPa Thrust Mode 3 (Vacuum Thrust Specified) Chamber Propellant Flow 502.9 kg/sec Sea Level Thrust 147600 kgf Vacuum Thrust 147600 kgf Atmospheric Thrust Loss 0 kgf Atmospheric Thrust Loss 0.0% Number of Engines on Stage 1 Estimated Ivac 293.5 sec Nozzle Stagnation Mach No 0.40 Injector Total Pressure 6.4 MPa Nozzle Inlet Total Pressure 5.8 MPa Nozzle Inlet Static Pressure 5.3 MPa Throat Static Pressure 3.2 MPa Nozzle Exit Static Pressure 10.9 kPa Chamber Temperature 3572.0°K Nozzle Inlet Temperature 3482.6°K Throat Temperature 3078.0°K Nozzle Exit Temperature 775.5°K Theoretical Characteristic Velocity 1705.7 m/sec Actual Characteristic Velocity 1663.1 m/sec Theoretical Vacuum Thrust Coefficient 1.761 Actual Vacuum Thrust Coefficient 1.731 Theoretical SL Thrust Coefficient 1.761 Actual SL Thrust Coefficient 1.731 Vacuum Exhaust Velocity 3003.3 m/sec Sea Level Exhaust Velocity 3003.3 m/sec Chamber Theoretical Vac Specific Impulse 306.3 sec Chamber Design Vac Specific Impulse 293.5 sec Chamber Theoretical SL Specific Impulse 306.3 sec Chamber Design SL Specific Impulse 293.5 sec Turbopump Propellant Usage 0.0% TurboPump Propellant Flow 0.0 kg/sec Engine Total Propellant Flow 502.9 kg/sec Engine Design SL Specific Impulse 293.5 sec

171

Appendix B2 Sample TRAJ2DF Data Files

Engine Design Vacuum Specific Impulse 293.5 sec SECOND STAGE ENGINE SIZING Chamber Length 0.548 m Chamber Diameter 0.540 m Length Converging Nozzle 0.212 m Chamber Total Length 0.760 m Throat Diameter 0.427 m Nozzle Bell Length 2.906 m Nozzle Exit Diameter 2.368 m Engine Total Length 3.666 m

FIRST STAGE ENGINE DESIGN Engine Viking VC Propellant N2O4/UH25 Chamber Pressure 5.8 MPa Gamma 1.335 Gas Molecular Weight 22.67 Combustion Temperature 3572°K Lstar 1.150 m Contraction Ratio 1.6 Expansion Ratio 12.2 etaVstar 0.975 etaF 0.983 Throat Half Angle 20.000 deg Nozzle Bell Percent 80.000 Ambient Operating Pressure 101.325 kPa Thrust Mode 2 (Sea Level Thrust Specified) Chamber Propellant Flow 509.0 kg/sec Sea Level Thrust 125000 kgf Vacuum Thrust 143174 kgf Atmospheric Thrust Loss 18174 kgf Atmospheric Thrust Loss 12.7% Number of Engines on Stage 5 Estimated Ivac 278.4 sec Nozzle Stagnation Mach No 0.40 Injector Total Pressure 6.4 MPa Nozzle Inlet Total Pressure 5.8 MPa Nozzle Inlet Static Pressure 5.3 MPa Throat Static Pressure 3.2 MPa Nozzle Exit Static Pressure 38.2 kPa Chamber Temperature 3572.0°K Nozzle Inlet Temperature 3479.0°K Throat Temperature 3059.5°K Nozzle Exit Temperature 1010.8°K Theoretical Characteristic Velocity 1699.4 m/sec Actual Characteristic Velocity 1656.9 m/sec Theoretical Vacuum Thrust Coefficient 1.690 Actual Vacuum Thrust Coefficient 1.665 Theoretical SL Thrust Coefficient 1.478 Actual SL Thrust Coefficient 1.453 Vacuum Exhaust Velocity 2871.7 m/sec Sea Level Exhaust Velocity 2512.6 m/sec Chamber Theoretical Vac Specific Impulse 292.8 sec Chamber Design Vac Specific Impulse 281.3 sec Chamber Theoretical SL Specific Impulse 256.2 sec Chamber Design SL Specific Impulse 245.6 sec Turbopump Propellant Usage 0.0% TurboPump Propellant Flow 0.0 kg/sec Engine Total Propellant Flow 509.0 kg/sec Engine Design SL Specific Impulse 245.6 sec Engine Design Vacuum Specific Impulse 281.3 sec

172

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Engine Design for Performance

Case Study: Hypothetical Launch Vehicle

FIRST STAGE ENGINE SIZING Chamber Length 0.548 m Chamber Diameter 0.542 m Length Converging Nozzle 0.213 m Chamber Total Length 0.760 m Throat Diameter 0.428 m Nozzle Bell Length 1.603 m Nozzle Exit Diameter 1.497 m Engine Total Length 2.363 m

173

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Launch Vehicle Stack

Case Study: Hypothetical Launch Vehicle

PAYLOAD AND ACCOMODATION Spacecraft Mass, kg 4768.0 Payload Adapter Mass, kg 200.0 Vehicle Equipment Bay Mass, kg 400.0 Total Payload Mass, kg 5368.0

THIRD STAGE Stage Mass, kg 30721.0 Vehicle Mass Ratio 3.9 Stage Delta V, m/sec 5891.0 Stage Structure, kg 3993.7 Stage Reserve Propellant, kg 0.0 Vehicle Mass @ Nominal Shutdown, kg 9361.7 Max Axial Load, g 1.6 Stage Nominal Propellant, kg 26727.3 Stage Nominal Burn Time, sec 793.2 Ignition Thrust/Weight 0.42 Vehicle Mass @ Ignition, kg 36089.0

SECOND STAGE Stage Mass, kg 91010.7 Vehicle Mass Ratio 2.7 Stage Delta V, m/sec 2897.6 Fairing Mass, kg 900.0 Stage Structure, kg 8191.0 Stage Reserve Propellant, kg 1597.2 Vehicle Mass @ Nominal Shutdown, kg 46777.1 Max Axial Load, g 3.3 Stage Nominal Propellant, kg 81222.6 Stage Nominal Burn Time, sec 161.5 Ignition Thrust/Weight 1.15 Vehicle Mass @ Ignition, kg 127999.7

FIRST STAGE Stage Mass, kg 344480.3 Vehicle Mass Ratio 3.1 Stage Delta V, m/sec 3126.1 Stage Structure, kg 24113.6 Stage Reserve Propellant, kg 0.0 Vehicle Mass @ Nominal Shutdown, kg 152113.3 Max Axial Load, g

174

Appendix B2 Sample TRAJ2DF Data Files

3.3 Stage Nominal Propellant, kg 320366.7 Stage Nominal Burn Time, sec 125.9 Ignition Thrust/Weight 1.32 Vehicle Mass @ Ignition, kg 472480.0

VEHICLE SUMMARY GLOW, kg 472480.0 Vehicle Total Delta-V, m/sec 11914.7 Vehicle Nominal Burn Time, sec 1080.6

175

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Mission Orbit Data

Case Study: Hypothetical Launch Vehicle

MISSION ORBITAL ELEMENTS REQUIRED Mission Mode Specified 0 S3 Shutdown Height, km 0.0 S3 Shutdown Velocity, m/sec 0.000 S3 Shutdown Flight Path Angle, deg 0.0 Orbit Semi-Major Axis, km 0.00 Orbit Eccentricity 0.0000 Orbit Argument of Perigee, deg 0.00 Orbit True Anomaly, deg 0.00 Orbit Perigee height, km 200.00 Orbit Apogee Radius, km 42164.50

176

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - First Stage Vertical Rise

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Stage Propellant | Pitch | Thrust | Acceleration | Velocity | FPA | Altitude| | | Mass | Used Remain Flow | | Total Loss Tx Ty | A Ax Ay | Mach V Vx Vy | | Y | | sec | kg | kg kg kg/sec | deg | Kgf Kgf Kgf Kgf | m/s2 m/s2 m/s2 | No m/s m/s m/s | deg | m | | 0.00 | 472480.0 | 0.0 320366.7 2545.2 | 90.00 | 624998.7 90872.3 0.0 624998.7 | 3.17 0.00 3.17 | 0.00 0.0 0.0 0.0 | 90.00 | 0.0 | | 1.00 | 469934.8 | 2545.2 317821.5 2545.2 | 90.00 | 625015.9 90855.1 0.0 625015.9 | 3.24 0.00 3.24 | 0.01 3.2 0.0 3.2 | 90.00 | 1.6 | | 2.00 | 467389.6 | 5090.4 315276.2 2545.2 | 90.00 | 625067.9 90803.1 0.0 625067.9 | 3.31 0.00 3.31 | 0.02 6.5 0.0 6.5 | 90.00 | 6.4 | | 3.00 | 464844.4 | 7635.6 312731.0 2545.2 | 90.00 | 625155.5 90715.5 0.0 625155.5 | 3.38 0.00 3.38 | 0.03 9.8 0.0 9.8 | 90.00 | 14.6 | | 4.00 | 462299.1 | 10180.9 310185.8 2545.2 | 90.00 | 625279.3 90591.7 0.0 625279.3 | 3.45 0.00 3.45 | 0.04 13.2 0.0 13.2 | 90.00 | 26.1 | | 5.00 | 459753.9 | 12726.1 307640.6 2545.2 | 90.00 | 625440.0 90431.0 0.0 625440.0 | 3.52 0.00 3.52 | 0.05 16.7 0.0 16.7 | 90.00 | 41.0 | | 6.00 | 457208.7 | 15271.3 305095.4 2545.2 | 90.00 | 625638.2 90232.9 0.0 625638.2 | 3.60 0.00 3.60 | 0.06 20.3 0.0 20.3 | 90.00 | 59.5 | | 7.00 | 454663.5 | 17816.5 302550.2 2545.2 | 90.00 | 625874.4 89996.6 0.0 625874.4 | 3.67 0.00 3.67 | 0.07 23.9 0.0 23.9 | 90.00 | 81.6 | | 7.20 | 454154.5 | 18325.5 302041.1 2545.2 | 90.00 | 625926.3 89944.7 0.0 625926.3 | 3.68 0.00 3.68 | 0.07 24.6 0.0 24.6 | 90.00 | 86.5 |

177

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - First Stage Pitch Over

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Stage Propellant | Pitch | Thrust | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | Used Remain Flow | | Total Loss Tx Ty | A Ax Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | kg kg kg/sec | deg | Kgf Kgf Kgf Kgf | m/s2 m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 7.20 | 454154.5 | 18325.5 302041.1 2545.2 | 90.00 | 625926.3 89944.7 0.0 625926.3 | 3.68 0.00 3.68 | 0.07 24.6 0.0 24.6 | 90.00 | .00 0.0 86.5 86.5 90.0 | | 8.21 | 451609.2 | 20870.8 299495.9 2545.2 | 89.84 | 626206.0 89665.0 1755.6 624686.7 | 3.76 0.04 3.76 | 0.08 28.4 0.0 28.4 | 89.94 | .00 0.0 112.9 112.9 90.0 | | 9.21 | 449064.0 | 23416.0 296950.7 2545.2 | 89.76 | 626527.5 89343.5 2589.3 624617.0 | 3.83 0.06 3.83 | 0.09 32.2 0.1 32.2 | 89.86 | .00 0.1 143.2 143.2 90.0 | | 10.21 | 446518.8 | 25961.2 294405.5 2545.2 | 89.67 | 626888.7 88982.4 3595.7 624544.6 | 3.91 0.08 3.91 | 0.11 36.0 0.1 36.0 | 89.77 | .00 0.2 177.3 177.3 89.9 | | 11.21 | 443973.6 | 28506.4 291860.3 2545.2 | 89.56 | 627289.9 88581.1 4783.3 624471.1 | 3.99 0.11 3.99 | 0.12 40.0 0.2 40.0 | 89.66 | .00 0.4 215.3 215.3 89.9 | | 12.21 | 441428.4 | 31051.6 289315.1 2545.2 | 89.43 | 627731.6 88139.4 6159.0 624398.1 | 4.07 0.14 4.06 | 0.13 44.0 0.4 44.0 | 89.53 | .00 0.7 257.3 257.3 89.9 | | 13.21 | 438883.2 | 33596.8 286769.8 2545.2 | 89.29 | 628214.0 87657.0 7728.1 624327.4 | 4.15 0.17 4.14 | 0.14 48.1 0.5 48.1 | 89.39 | .00 1.1 303.3 303.3 89.8 | | 14.21 | 436338.0 | 36142.0 284224.6 2545.2 | 89.13 | 628737.5 87133.5 9494.8 624260.7 | 4.23 0.21 4.22 | 0.15 52.3 0.7 52.3 | 89.23 | .00 1.7 353.5 353.5 89.7 | | 15.21 | 433792.7 | 38687.3 281679.4 2545.2 | 88.95 | 629302.1 86568.9 11462.1 624199.8 | 4.31 0.26 4.30 | 0.17 56.6 0.9 56.6 | 89.05 | .00 2.5 407.9 407.9 89.6 | | 16.21 | 431247.5 | 41232.5 279134.2 2545.2 | 88.75 | 629908.1 85962.9 13631.7 624146.4 | 4.40 0.31 4.39 | 0.18 60.9 1.2 60.9 | 88.85 | .00 3.6 466.6 466.6 89.6 | | 17.21 | 428702.3 | 43777.7 276589.0 2545.2 | 88.53 | 630555.6 85315.4 16004.7 624102.2 | 4.49 0.37 4.47 | 0.19 65.3 1.6 65.3 | 88.63 | .00 5.0 529.7 529.8 89.5 | | 18.21 | 426157.1 | 46322.9 274043.8 2545.2 | 88.29 | 631244.6 84626.4 18581.0 624068.7 | 4.58 0.43 4.56 | 0.21 69.9 2.0 69.8 | 88.39 | .00 6.7 597.3 597.3 89.4 | | 19.21 | 423611.9 | 48868.1 271498.6 2545.2 | 88.04 | 631975.0 83896.0 21360.0 624047.5 | 4.67 0.49 4.64 | 0.22 74.5 2.4 74.4 | 88.14 | .00 8.9 669.4 669.5 89.2 | | 20.21 | 421066.7 | 51413.3 268953.3 2545.2 | 87.77 | 632746.9 83124.2 24340.3 624040.0 | 4.76 0.57 4.73 | 0.23 79.2 3.0 79.1 | 87.86 | .00 11.6 746.2 746.3 89.1 | | 21.21 | 418521.4 | 53958.6 266408.1 2545.2 | 87.47 | 633559.9 82311.1 27519.7 624047.1 | 4.86 0.64 4.82 | 0.25 84.0 3.6 83.9 | 87.57 | .00 14.9 827.7 827.9 89.0 | | 22.21 | 415976.2 | 56503.8 263862.9 2545.2 | 87.17 | 634414.0 81457.0 30895.9 624069.9 | 4.96 0.73 4.91 | 0.26 88.9 4.2 88.8 | 87.26 | .00 18.8 914.1 914.2 88.8 | | 23.21 | 413431.0 | 59049.0 261317.7 2545.2 | 86.84 | 635308.9 80562.1 34465.7 624108.7 | 5.07 0.82 5.00 | 0.28 93.9 5.0 93.7 | 86.94 | .00 23.4 1005.3 1005.6 88.7 | | 24.21 | 410885.8 | 61594.2 258772.5 2545.2 | 86.50 | 636244.1 79626.9 38225.8 624163.6 | 5.17 0.91 5.09 | 0.29 98.9 5.9 98.8 | 86.59 | .00 28.8 1101.5 1101.9 88.5 | | 25.21 | 408340.6 | 64139.4 256227.3 2545.2 | 86.14 | 637219.3 78651.7 42172.1 624234.2 | 5.29 1.01 5.19 | 0.31 104.1 6.8103.9 | 86.23 | .00 35.2 1202.8 1203.4 88.3 | | 26.21 | 405795.4 | 66684.6 253682.1 2545.2 | 85.76 | 638234.0 77637.0 46300.4 624319.7 | 5.40 1.12 5.28 | 0.33 109.4 7.9109.1 | 85.85 | .00 42.5 1309.4 1310.0 88.1 | | 27.21 | 403250.2 | 69229.8 251136.8 2545.2 | 85.37 | 639287.5 76583.5 50606.3 624419.3 | 5.52 1.23 5.38 | 0.34 114.8 9.1114.5 | 85.46 | .00 51.0 1421.2 1422.1 87.9 | | 28.21 | 400704.9 | 71775.1 248591.6 2545.2 | 84.96 | 640379.3 75491.7 55084.9 624531.4 | 5.64 1.35 5.48 | 0.36 120.410.4119.9 | 85.06 | .00 60.7 1538.3 1539.5 87.7 | | 29.21 | 398159.7 | 74320.3 246046.4 2545.2 | 84.54 | 641508.7 74362.4 59730.9 624654.0 | 5.77 1.47 5.58 | 0.38 126.011.8125.4 | 84.63 | .00 71.8 1661.0 1662.6 87.5 | | 30.21 | 395614.5 | 76865.5 243501.2 2545.2 | 84.10 | 642674.7 73196.3 64539.0 624784.8 | 5.91 1.60 5.69 | 0.40 131.713.3131.1 | 84.20 | .00 84.3 1789.2 1791.2 87.3 | | 31.21 | 393069.3 | 79410.7 240956.0 2545.2 | 83.65 | 643876.5 71994.5 69503.7 624922.3 | 6.04 1.73 5.79 | 0.41 137.615.0136.8 | 83.75 | .00 98.5 1923.2 1925.7 87.1 | | 32.21 | 390524.1 | 81955.9 238410.8 2545.2 | 83.19 | 645113.2 70757.8 74619.3 625064.7 | 6.19 1.87 5.89 | 0.43 143.616.8142.7 | 83.29 | .00 114.3 2062.9 2066.1 86.8 | | 33.21 | 387978.9 | 84501.1 235865.5 2545.2 | 82.72 | 646383.6 69487.4 79879.9 625210.2 | 6.33 2.02 6.00 | 0.45 149.818.7148.6 | 82.81 | .00 132.1 2208.5 2212.5 86.6 | | 34.21 | 385433.7 | 87046.3 233320.3 2545.2 | 82.23 | 647686.7 68184.3 85279.6 625357.1 | 6.48 2.17 6.11 | 0.47 156.120.8154.7 | 82.33 | .00 151.9 2360.1 2365.0 86.3 | | 35.21 | 382888.4 | 89591.6 230775.1 2545.2 | 81.74 | 649021.2 66849.8 90812.4 625503.6 | 6.64 2.33 6.22 | 0.49 162.523.1160.8 | 81.83 | .00 173.8 2517.9 2523.9 86.1 | | 36.21 | 380343.2 | 92136.8 228229.9 2545.2 | 81.23 | 650385.7 65485.3 96472.1 625648.3 | 6.80 2.49 6.33 | 0.51 169.025.5167.1 | 81.33 | .00 198.0 2681.8 2689.1 85.8 | | 37.21 | 377798.0 | 94682.0 225684.7 2545.2 | 80.72 | 651779.0 64092.0 102252.8 625790.0 | 6.97 2.65 6.44 | 0.53 175.728.1173.5 | 80.81 | .00 224.8 2852.1 2861.0 85.5 | | 38.21 | 375252.8 | 97227.2 223139.5 2545.2 | 80.20 | 653199.4 62671.6 108148.5 625928.1 | 7.14 2.83 6.56 | 0.56 182.630.8180.0 | 80.29 | .00 254.2 3028.8 3039.5 85.2 | | 39.21 | 372707.6 | 99772.4 220594.3 2545.2 | 79.67 | 654645.4 61225.6 114153.2 626062.7 | 7.32 3.00 6.67 | 0.58 189.633.7186.6 | 79.76 | .00 286.4 3212.1 3224.9 84.9 | | 39.70 | 371460.4 | ******* 219347.1 2545.2 | 79.50 | 655362.8 60508.2 116082.2 626323.4 | 7.40 3.06 6.74 | 0.59 193.135.2189.9 | 79.50 | .00 303.3 3304.4 3318.3 84.8 |

178

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - First Stage Gravity Turn to Burnout

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Stage Propellant | Pitch | Thrust | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | Used Remain Flow | | Total Loss Tx Ty | A Ax Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | kg kg kg/sec | deg | Kgf Kgf Kgf Kgf | m/s2 m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 39.70 | 371460.4 | 101019.6 219347.1 2545.2 | 79.50 | 655377.5 60493.5 116082.2 626323.4 | 7.40 3.06 6.74 | 0.59 193.1 35.2 189.9 | 79.50 | 0.00 303.3 3304.4 3318.3 84.8 | | 40.70 | 368915.2 | 103564.8 216801.9 2545.2 | 78.97 | 656844.0 59027.0 122093.1 626493.5 | 7.59 3.25 6.86 | 0.61 200.4 38.4 196.7 | 78.97 | 0.00 340.0 3497.7 3514.2 84.4 | | 41.70 | 366370.0 | 106110.0 214256.7 2545.2 | 78.43 | 658346.5 57524.5 128247.1 626658.3 | 7.78 3.43 6.98 | 0.64 207.8 41.7 203.6 | 78.43 | 0.00 380.0 3697.8 3717.3 84.1 | | 42.70 | 363824.8 | 108655.2 211711.5 2545.2 | 77.89 | 659868.6 56002.5 134490.3 626830.3 | 7.97 3.62 7.10 | 0.66 215.4 45.2 210.7 | 77.88 | 0.00 423.4 3904.9 3927.8 83.8 | | 43.70 | 361279.6 | 111200.4 209166.2 2545.2 | 77.34 | 661408.2 54462.8 140813.3 626991.2 | 8.17 3.82 7.22 | 0.69 223.2 48.9 217.8 | 77.34 | 0.00 470.4 4119.2 4145.9 83.5 | | 44.70 | 358734.4 | 113745.6 206621.0 2545.2 | 76.79 | 662963.5 52907.5 147198.8 627090.4 | 8.38 4.02 7.35 | 0.72 231.2 52.9 225.1 | 76.78 | 0.00 521.3 4340.6 4371.8 83.1 | | 45.70 | 356189.1 | 116290.9 204075.8 2545.2 | 76.23 | 664532.4 51338.6 153622.6 627058.4 | 8.58 4.23 7.47 | 0.74 239.4 57.0 232.5 | 76.23 | 0.01 576.2 4569.4 4605.6 82.8 | | 46.70 | 353643.9 | 118836.1 201530.6 2545.2 | 75.68 | 666112.6 49758.4 160052.6 626805.2 | 8.79 4.44 7.59 | 0.77 247.8 61.3 240.0 | 75.67 | 0.01 635.2 4805.7 4847.5 82.5 | | 47.70 | 351098.7 | 121381.3 198985.4 2545.2 | 75.12 | 667701.9 48169.1 166447.1 626216.9 | 8.99 4.65 7.70 | 0.80 256.3 65.9 247.7 | 75.11 | 0.01 698.8 5049.5 5097.7 82.1 | | 48.70 | 348553.5 | 123926.5 196440.2 2545.2 | 74.55 | 669298.0 46573.1 172741.1 625113.5 | 9.19 4.86 7.80 | 0.83 265.0 70.6 255.4 | 74.55 | 0.01 766.9 5301.1 5356.3 81.8 | | 49.70 | 346008.3 | 126471.7 193895.0 2545.2 | 73.99 | 670898.2 44972.8 178800.2 623097.3 | 9.36 5.07 7.87 | 0.86 273.9 75.6 263.3 | 73.98 | 0.01 840.0 5560.4 5623.6 81.4 | | 50.70 | 343463.1 | 129016.9 191349.7 2545.2 | 73.42 | 672500.1 43370.9 184433.1 619635.0 | 9.50 5.27 7.90 | 0.89 282.9 80.7 271.2 | 73.42 | 0.01 918.0 5827.6 5899.6 81.0 | | 51.70 | 340917.9 | 131562.1 188804.5 2545.2 | 72.86 | 674100.6 41770.4 189433.5 614217.5 | 9.58 5.45 7.88 | 0.92 292.0 86.1 279.1 | 72.85 | 0.01 1001.4 6102.8 6184.4 80.7 | | 52.70 | 338372.6 | 134107.4 186259.3 2545.2 | 72.29 | 675696.5 40174.5 193883.1 607303.1 | 9.62 5.62 7.81 | 0.96 301.2 91.6 286.9 | 72.29 | 0.01 1090.1 6385.8 6478.2 80.3 | | 53.70 | 335827.4 | 136652.6 183714.1 2545.2 | 71.73 | 677284.6 38586.4 198112.0 600042.9 | 9.66 5.78 7.73 | 0.99 310.3 97.3 294.7 | 71.72 | 0.01 1184.5 6676.6 6780.9 79.9 | | 54.70 | 333282.2 | 139197.8 181168.9 2545.2 | 71.16 | 678861.5 37009.5 202540.9 593705.4 | 9.72 5.96 7.68 | 1.02 319.5 103.2 302.4 | 71.16 | 0.01 1284.6 6975.1 7092.5 79.6 | | 55.70 | 330737.0 | 141743.0 178623.7 2545.2 | 70.60 | 680424.3 35446.8 207375.6 588792.6 | 9.83 6.15 7.67 | 1.06 328.7 109.2 310.1 | 70.59 | 0.01 1390.7 7281.3 7413.1 79.2 | | 56.70 | 328191.8 | 144288.2 176078.5 2545.2 | 70.03 | 681970.2 33900.8 212638.1 585239.5 | 9.98 6.35 7.70 | 1.09 338.1 115.5 317.8 | 70.03 | 0.01 1502.9 7595.2 7742.7 78.8 | | 57.70 | 325646.6 | 146833.4 173533.2 2545.2 | 69.47 | 683496.8 32374.2 218332.3 582941.3 | 10.18 6.57 7.77 | 1.13 347.6 121.9 325.5 | 69.46 | 0.01 1621.5 7916.9 8081.4 78.4 | | 58.70 | 323101.3 | 149378.7 170988.0 2545.2 | 68.90 | 685002.0 30869.0 224451.7 581778.9 | 10.41 6.81 7.88 | 1.16 357.3 128.6 333.3 | 68.90 | 0.02 1746.6 8246.2 8429.4 78.0 | | 59.70 | 320556.1 | 151923.9 168442.8 2545.2 | 68.34 | 686483.9 29387.1 230990.8 581649.9 | 10.68 7.07 8.01 | 1.20 367.2 135.6 341.3 | 68.33 | 0.02 1878.5 8583.5 8786.9 77.6 | | 60.70 | 318010.9 | 154469.1 165897.6 2545.2 | 67.78 | 687940.6 27930.4 237925.3 582417.1 | 10.99 7.34 8.18 | 1.24 377.4 142.8 349.4 | 67.77 | 0.02 2017.4 8928.8 9154.2 77.3 | | 61.70 | 315465.7 | 157014.3 163352.4 2545.2 | 67.22 | 689370.6 26500.4 245200.7 583886.7 | 11.32 7.62 8.37 | 1.28 387.9 150.2 357.6 | 67.21 | 0.02 2163.7 9282.3 9531.5 76.9 | | 62.70 | 312920.5 | 159559.5 160807.2 2545.2 | 66.66 | 690772.3 25098.7 252723.6 585796.0 | 11.68 7.92 8.58 | 1.32 398.7 158.0 366.1 | 66.66 | 0.02 2317.5 9644.2 9919.1 76.5 | | 63.70 | 310375.3 | 162104.7 158262.0 2545.2 | 66.11 | 692144.3 23726.7 260392.8 587892.0 | 12.04 8.23 8.80 | 1.37 409.9 166.1 374.8 | 66.10 | 0.02 2479.3 10014.6 10317.4 76.1 | | 64.70 | 307830.1 | 164649.9 155716.7 2545.2 | 65.56 | 693485.2 22385.8 268100.7 589933.1 | 12.42 8.54 9.02 | 1.41 421.5 174.4 383.7 | 65.55 | 0.02 2649.2 10393.8 10726.7 75.7 | | 65.70 | 305284.8 | 167195.2 153171.5 2545.2 | 65.01 | 694793.6 21077.4 275697.2 591610.5 | 12.79 8.86 9.23 | 1.46 433.4 183.1 392.9 | 65.01 | 0.03 2827.7 10782.1 11147.3 75.3 | MaxQ: Time= 66.31sec, Altitude=11019m, 36153feet, MaxQ= 3604 kgf/m^2, 738 lbf/ft^2, Mach No=1.49 | 66.70 | 302739.6 | 169740.4 150626.3 2545.2 | 64.47 | 696067.0 19804.0 283035.9 592652.4 | 13.15 9.17 9.42 | 1.51 445.7 192.1 402.2 | 64.47 | 0.03 3015.0 11179.6 11579.7 74.9 | | 67.70 | 300194.4 | 172285.6 148081.1 2545.2 | 63.93 | 697293.4 18577.6 290160.1 593199.8 | 13.49 9.48 9.61 | 1.55 458.3 201.4 411.7 | 63.93 | 0.03 3211.4 11586.5 12024.1 74.5 | | 68.70 | 297649.2 | 174830.8 145535.9 2545.2 | 63.40 | 698469.9 17401.1 297142.5 593438.8 | 13.84 9.79 9.78 | 1.60 471.3 211.1 421.4 | 63.40 | 0.03 3417.2 12003.1 12480.9 74.1 | | 69.70 | 295104.0 | 177376.0 142990.7 2545.2 | 62.88 | 699596.9 16274.1 304041.7 593510.6 | 14.18 10.10 9.95 | 1.64 484.6 221.0 431.3 | 62.87 | 0.03 3632.8 12429.4 12950.4 73.7 | | 70.70 | 292558.8 | 179921.2 140445.4 2545.2 | 62.35 | 700674.5 15196.5 310866.0 593447.8 | 14.53 10.42 10.12 | 1.69 498.2 231.2 441.3 | 62.35 | 0.03 3858.4 12865.7 13432.9 73.3 | | 71.70 | 290013.6 | 182466.4 137900.2 2545.2 | 61.84 | 701703.2 14167.8 317606.7 593247.8 | 14.87 10.74 10.29 | 1.74 512.2 241.8 451.5 | 61.83 | 0.04 4094.4 13312.1 13928.8 72.9 | | 72.70 | 287468.3 | 185011.7 135355.0 2545.2 | 61.33 | 702683.4 13187.6 324255.6 592909.2 | 15.22 11.06 10.46 | 1.78 526.5 252.7 461.9 | 61.32 | 0.04 4341.1 13768.9 14438.4 72.5 | | 73.70 | 284923.1 | 187556.9 132809.8 2545.2 | 60.82 | 703615.9 12255.1 330805.4 592431.5 | 15.57 11.38 10.63 | 1.83 541.2 263.9 472.5 | 60.82 | 0.04 4598.8 14236.1 14962.0 72.1 | | 74.70 | 282377.9 | 190102.1 130264.6 2545.2 | 60.32 | 704501.3 11369.7 337249.6 591815.7 | 15.92 11.71 10.79 | 1.88 556.2 275.4 483.2 | 60.32 | 0.04 4867.8 14713.9 15500.0 71.7 |

Ú | 115.70 | 178024.1 | 294455.9 25910.8 2545.2 | 45.30 | 715744.8 126.3 503395.2 508719.6 | 33.25 27.73 18.35 | 4.60 1501.5 1056.1 1067.2 | 45.30 | 0.27 29856.6 45461.0 54447.0 56.5 | | 116.70 | 175478.9 | 297001.1 23365.6 2545.2 | 45.05 | 715760.8 110.2 505608.9 506557.5 | 33.85 28.25 18.64 | 4.67 1534.3 1084.0 1085.9 | 45.05 | 0.28 30919.0 46537.5 55934.8 56.2 | | 117.70 | 172933.7 | 299546.3 20820.4 2545.2 | 44.81 | 715774.9 96.1 507768.6 504425.0 | 34.46 28.79 18.94 | 4.75 1567.8 1112.3 1104.9 | 44.81 | 0.29 32009.0 47632.9 57455.4 55.9 | | 118.70 | 170388.5 | 302091.5 18275.1 2545.2 | 44.57 | 715787.3 83.7 509876.0 502322.9 | 35.09 29.34 19.25 | 4.86 1601.9 1141.2 1124.2 | 44.57 | 0.30 33127.2 48747.4 59009.4 55.6 | | 119.70 | 167843.2 | 304636.8 15729.9 2545.2 | 44.34 | 715798.3 72.7 511932.2 500251.3 | 35.74 29.91 19.57 | 4.96 1636.6 1170.6 1143.8 | 44.34 | 0.31 34274.2 49881.3 60597.5 55.2 | | 120.70 | 165298.0 | 307182.0 13184.7 2545.2 | 44.11 | 715808.0 63.0 513938.6 498210.8 | 36.41 30.49 19.90 | 5.07 1672.0 1200.6 1163.7 | 44.11 | 0.32 35450.4 51035.1 62220.3 54.9 | | 121.70 | 162752.8 | 309727.2 10639.5 2545.2 | 43.89 | 715816.6 54.4 515896.6 496201.9 | 37.09 31.08 20.25 | 5.20 1708.1 1231.1 1184.0 | 43.88 | 0.33 36656.4 52208.9 63878.5 54.7 | | 122.70 | 160207.6 | 312272.4 8094.3 2545.2 | 43.67 | 715824.2 46.8 517807.2 494224.7 | 37.80 31.69 20.60 | 5.34 1744.9 1262.3 1204.7 | 43.66 | 0.34 37892.9 53403.3 65572.8 54.4 | | 123.70 | 157662.4 | 314817.6 5549.1 2545.2 | 43.45 | 715830.9 40.1 519671.3 492279.2 | 38.53 32.32 20.97 | 5.49 1782.4 1294.1 1225.7 | 43.45 | 0.35 39160.2 54618.5 67304.0 54.1 | | 124.70 | 155117.2 | 317362.8 3003.9 2545.2 | 43.24 | 715836.8 34.2 521489.9 490365.4 | 39.28 32.96 21.36 | 5.65 1820.7 1326.4 1247.2 | 43.24 | 0.36 40459.2 55854.9 69072.8 53.8 | | 125.70 | 152572.0 | 319908.0 458.6 2545.2 | 43.03 | 715842.0 29.0 523263.9 488483.5 | 40.05 33.63 21.76 | 5.81 1859.7 1359.5 1269.0 | 43.03 | 0.38 41790.3 57113.0 70879.8 53.5 | | 125.88 | 152113.3 | 320366.7 0.0 2545.2 | 42.99 | 715842.9 28.1 523596.0 488129.5 | 40.20 33.76 21.84 | 5.84 1866.8 1365.5 1273.0 | 42.99 | 0.38 42033.6 57342.0 71209.6 53.4 | DynP@S1BO = 76 kgf/m^2, 15 lbf/ft^2 179

Appendix B2 Sample TRAJ2DF Data Files

Traj2d - First Stage Coast to Impact

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | A Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 125.88 | 24113.6 | -9.6 -9.6 | 5.84 1866.8 1365.50 1273.0 | 42.99 | 0.38 42033.6 57342.0 71209.6 53.4 | | 126.87 | 24113.6 | -9.6 -9.6 | 5.84 1860.3 1365.23 1263.7 | 42.79 | 0.39 43386.7 58610.3 73040.2 53.2 | | 127.87 | 24113.6 | -9.6 -9.6 | 5.84 1853.8 1364.96 1254.3 | 42.58 | 0.40 44739.2 59869.3 74864.7 52.9 | | 128.87 | 24113.6 | -9.6 -9.6 | 5.84 1847.3 1364.70 1245.0 | 42.37 | 0.41 46091.2 61119.0 76683.0 52.6 | | 129.87 | 24113.6 | -9.6 -9.6 | 5.84 1840.8 1364.44 1235.7 | 42.16 | 0.43 47442.7 62359.3 78495.2 52.4 | | 130.87 | 24113.6 | -9.6 -9.6 | 5.84 1834.4 1364.18 1226.3 | 41.95 | 0.44 48793.7 63590.3 80301.2 52.1 | | 131.87 | 24113.6 | -9.6 -9.6 | 5.84 1827.9 1363.92 1217.0 | 41.74 | 0.45 50144.2 64812.0 82101.0 51.9 | | 132.87 | 24113.6 | -9.6 -9.6 | 5.84 1821.6 1363.66 1207.7 | 41.53 | 0.46 51494.1 66024.3 83894.5 51.7 | | 133.87 | 24113.6 | -9.6 -9.6 | 5.84 1815.2 1363.41 1198.4 | 41.31 | 0.47 52843.5 67227.4 85681.8 51.4 | | 134.87 | 24113.6 | -9.6 -9.6 | 5.84 1808.9 1363.15 1189.1 | 41.10 | 0.49 54192.5 68421.1 87462.9 51.2 | | 135.87 | 24113.6 | -9.6 -9.6 | 5.84 1802.6 1362.90 1179.8 | 40.88 | 0.50 55540.9 69605.5 89237.7 51.0 | | 136.87 | 24113.6 | -9.6 -9.6 | 5.84 1796.3 1362.65 1170.4 | 40.66 | 0.51 56888.8 70780.6 91006.2 50.8 | | 137.87 | 24113.6 | -9.6 -9.6 | 5.84 1790.1 1362.41 1161.1 | 40.44 | 0.52 58236.3 71946.4 92768.4 50.6 | | 138.87 | 24113.6 | -9.6 -9.6 | 5.84 1783.9 1362.16 1151.8 | 40.22 | 0.54 59583.3 73102.9 94524.4 50.4 | | 139.87 | 24113.6 | -9.6 -9.6 | 5.84 1777.7 1361.92 1142.5 | 39.99 | 0.55 60929.8 74250.1 96274.1 50.2 | | 140.87 | 24113.6 | -9.6 -9.6 | 5.84 1771.6 1361.68 1133.3 | 39.77 | 0.56 62275.8 75388.0 98017.5 50.0 | | 141.87 | 24113.6 | -9.6 -9.6 | 5.84 1765.5 1361.44 1124.0 | 39.54 | 0.57 63621.3 76516.6 99754.6 49.8 | | 142.87 | 24113.6 | -9.6 -9.6 | 5.84 1759.4 1361.21 1114.7 | 39.31 | 0.58 64966.4 77635.9 101485.5 49.6 | | 143.87 | 24113.6 | -9.6 -9.6 | 5.84 1753.3 1360.97 1105.4 | 39.08 | 0.60 66311.0 78745.9 103210.1 49.4 | | 144.87 | 24113.6 | -9.6 -9.6 | 5.84 1747.3 1360.74 1096.1 | 38.85 | 0.61 67655.1 79846.7 104928.4 49.2 |

Ú | 413.87 | 24113.6 | -9.7 -9.7 | 5.84 1938.0 1368.48 -1372.2 | -45.08 | 3.82 425092.8 43328.8 428650.8 3.9 | | 414.87 | 24113.6 | -9.7 -9.7 | 5.84 1944.8 1368.78 -1381.6 | -45.27 | 3.83 426452.4 41951.9 429825.1 3.7 | | 415.87 | 24113.6 | -9.7 -9.7 | 5.84 1951.7 1369.07 -1391.0 | -45.45 | 3.84 427812.5 40565.7 431003.6 3.5 | | 416.87 | 24113.6 | -9.7 -9.7 | 5.84 1958.6 1369.37 -1400.4 | -45.64 | 3.86 429173.2 39170.0 432186.5 3.3 | | 417.87 | 24113.6 | -9.7 -9.7 | 5.84 1965.6 1369.67 -1409.8 | -45.83 | 3.87 430534.5 37764.9 433373.8 3.1 | | 418.87 | 24113.6 | -9.7 -9.7 | 5.84 1972.5 1369.97 -1419.2 | -46.01 | 3.88 431896.4 36350.4 434565.6 2.9 | | 419.87 | 24113.6 | -9.7 -9.7 | 5.84 1979.5 1370.28 -1428.6 | -46.19 | 3.89 433258.9 34926.6 435762.0 2.6 | | 420.87 | 24113.6 | -9.7 -9.7 | 5.84 1986.5 1370.58 -1438.0 | -46.37 | 3.90 434622.0 33493.3 436962.9 2.4 | | 421.87 | 24113.6 | -9.7 -9.7 | 5.84 1993.6 1370.89 -1447.4 | -46.56 | 3.92 435985.8 32050.6 438168.5 2.2 | | 422.87 | 24113.6 | -9.7 -9.7 | 5.84 2000.6 1371.20 -1456.8 | -46.73 | 3.93 437350.1 30598.5 439378.8 2.0 | | 423.87 | 24113.6 | -9.7 -9.7 | 5.84 2007.7 1371.51 -1466.2 | -46.91 | 3.94 438715.1 29137.0 440593.8 1.8 | | 424.87 | 24113.6 | -9.7 -9.7 | 5.84 2014.8 1371.83 -1475.7 | -47.09 | 3.95 440080.7 27666.0 441813.7 1.6 | | 425.87 | 24113.6 | -9.7 -9.7 | 5.84 2022.0 1372.15 -1485.1 | -47.26 | 3.97 441446.9 26185.6 443038.4 1.4 | | 426.87 | 24113.6 | -9.7 -9.7 | 5.84 2029.1 1372.46 -1494.5 | -47.44 | 3.98 442813.7 24695.8 444268.0 1.2 | | 427.87 | 24113.6 | -9.7 -9.7 | 5.84 2036.3 1372.79 -1504.0 | -47.61 | 3.99 444181.2 23196.6 445502.6 1.0 | | 428.87 | 24113.6 | -9.7 -9.7 | 5.84 2043.5 1373.11 -1513.4 | -47.78 | 4.00 445549.4 21687.9 446742.2 0.8 | | 429.87 | 24113.6 | -9.7 -9.7 | 5.84 2050.7 1373.44 -1522.9 | -47.95 | 4.01 446918.2 20169.8 447987.0 0.6 | | 430.87 | 24113.6 | -9.7 -9.7 | 5.84 2058.0 1373.76 -1532.3 | -48.12 | 4.03 448287.6 18642.2 449236.8 0.4 | | 431.87 | 24113.6 | -9.8 -9.8 | 5.84 2065.2 1374.09 -1541.8 | -48.29 | 4.04 449657.7 17105.1 450491.8 0.2 | | 432.87 | 24113.6 | -9.8 -9.8 | 5.84 2072.5 1374.43 -1551.2 | -48.46 | 4.05 451028.4 15558.6 451752.1 -0.1 | | 433.87 | 24113.6 | -9.8 -9.8 | 5.84 2079.8 1374.76 -1560.7 | -48.62 | 4.06 452399.8 14002.7 453017.7 -0.3 | | 434.87 | 24113.6 | -9.8 -9.8 | 5.84 2087.2 1375.10 -1570.2 | -48.79 | 4.08 453771.9 12437.2 454288.6 -0.5 | | 435.87 | 24113.6 | -9.8 -9.8 | 5.84 2094.5 1375.44 -1579.6 | -48.95 | 4.09 455144.7 10862.3 455565.0 -0.7 | | 436.87 | 24113.6 | -9.8 -9.8 | 5.84 2101.9 1375.78 -1589.1 | -49.12 | 4.10 456518.1 9278.0 456846.7 -0.9 | | 437.87 | 24113.6 | -9.8 -9.8 | 5.84 2109.3 1376.12 -1598.6 | -49.28 | 4.11 457892.2 7684.1 458134.0 -1.1 | | 438.87 | 24113.6 | -9.8 -9.8 | 5.84 2116.7 1376.47 -1608.1 | -49.44 | 4.13 459267.1 6080.8 459426.9 -1.3 | | 439.87 | 24113.6 | -9.8 -9.8 | 5.84 2124.2 1376.81 -1617.6 | -49.60 | 4.14 460642.6 4467.9 460725.4 -1.5 | | 440.87 | 24113.6 | -9.8 -9.8 | 5.84 2131.7 1377.16 -1627.1 | -49.76 | 4.15 462018.8 2845.6 462029.5 -1.7 | | 441.87 | 24113.6 | -9.8 -9.8 | 5.84 2139.2 1377.52 -1636.6 | -49.91 | 4.16 463395.7 1213.7 463339.4 -1.9 | | 442.61 | 24113.6 | -9.8 -9.8 | 5.84 2144.7 1377.78 -1643.6 | -50.03 | 4.17 464415.1 0.0 464312.5 -2.1 | 180

Appendix B2 Sample TRAJ2DF Data Files

Traj2d - Stage 1/2 Interstage Coast

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | A Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 125.88 | 152113.3 | -9.6 -9.6 | 5.84 1866.8 1365.50 1273.0 | 42.99 | 0.38 42033.6 57342.0 71209.6 53.4 | | 126.87 | 152113.3 | -9.6 -9.6 | 5.86 1860.3 1365.23 1263.7 | 42.79 | 0.39 43386.7 58610.3 73040.2 53.2 | | 127.87 | 152113.3 | -9.6 -9.6 | 5.88 1853.8 1364.96 1254.3 | 42.58 | 0.40 44739.2 59869.3 74864.7 52.9 | | 128.87 | 152113.3 | -9.6 -9.6 | 5.90 1847.3 1364.70 1245.0 | 42.37 | 0.41 46091.2 61119.0 76683.0 52.6 | | 129.87 | 152113.3 | -9.6 -9.6 | 5.92 1840.8 1364.44 1235.7 | 42.16 | 0.43 47442.7 62359.3 78495.2 52.4 | | 130.87 | 152113.3 | -9.6 -9.6 | 5.94 1834.4 1364.18 1226.3 | 41.95 | 0.44 48793.7 63590.3 80301.2 52.1 | | 130.88 | 152113.3 | -9.6 -9.6 | 5.94 1834.3 1364.17 1226.2 | 41.95 | 0.44 48807.2 63602.6 80319.2 52.1 |

181

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Second Stage Powered Flight

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Stage Propellant | Pitch | Thrust | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | Used Remain Flow | | Total Loss Tx Ty | A Ax Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | kg kg kg/sec | deg | Kgf Kgf Kgf Kgf | m/s2 m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 130.88 | 127999.7 | 0.0 81222.6 502.9 | 48.10 | 147599.7 0.0 98571.9 109860.2 | 7.65 7.55 -1.20 | 5.94 1834.3 1824.5 1226.2 | 41.95 | 0.44 48807.2 63602.6 80319.2 52.1 | | 131.88 | 127496.8 | 502.9 80719.7 502.9 | 47.98 | 147599.7 0.0 98808.2 109647.6 | 7.69 7.60 -1.18 | 5.94 2204.0 1831.8 1225.6 | 33.79 | 0.45 50617.1 64828.5 82406.6 51.6 | | 132.88 | 126994.0 | 1005.7 80216.9 502.9 | 47.85 | 147599.7 0.0 99045.2 109433.6 | 7.74 7.65 -1.16 | 5.94 2209.7 1839.0 1224.9 | 33.67 | 0.47 52433.9 66053.7 84503.6 51.2 | | 133.88 | 126491.1 | 1508.6 79714.0 502.9 | 47.73 | 147599.7 0.0 99282.9 109218.0 | 7.78 7.70 -1.14 | 5.94 2215.4 1846.4 1224.3 | 33.55 | 0.49 54257.6 67278.3 86610.0 50.7 | | 134.88 | 125988.3 | 2011.4 79211.1 502.9 | 47.60 | 147599.7 0.0 99521.3 109000.9 | 7.83 7.75 -1.11 | 5.94 2221.2 1853.7 1223.7 | 33.43 | 0.50 56088.1 68502.4 88725.5 50.3 | | 135.88 | 125485.4 | 2514.3 78708.3 502.9 | 47.48 | 147599.7 0.0 99760.3 108782.1 | 7.87 7.80 -1.09 | 5.94 2227.1 1861.2 1223.1 | 33.31 | 0.52 57925.7 69725.8 90850.0 49.9 | | 136.88 | 124982.6 | 3017.1 78205.4 502.9 | 47.35 | 147599.7 0.0 100000.0 108561.9 | 7.92 7.85 -1.07 | 5.94 2233.1 1868.6 1222.6 | 33.20 | 0.54 59770.2 70948.7 92983.3 49.5 | | 137.88 | 124479.7 | 3520.0 77702.6 502.9 | 47.22 | 147599.7 0.0 100240.3 108340.0 | 7.97 7.90 -1.05 | 5.94 2239.1 1876.1 1222.1 | 33.08 | 0.55 61621.8 72171.0 95125.4 49.1 | | 138.88 | 123976.8 | 4022.9 77199.7 502.9 | 47.10 | 147599.7 0.0 100481.3 108116.5 | 8.01 7.95 -1.03 | 5.94 2245.1 1883.7 1221.6 | 32.96 | 0.57 63480.5 73392.8 97276.0 48.7 | | 139.88 | 123474.0 | 4525.7 76696.8 502.9 | 46.97 | 147599.7 0.0 100723.0 107891.4 | 8.06 8.00 -1.01 | 5.94 2251.3 1891.3 1221.1 | 32.85 | 0.59 65346.4 74614.2 99435.1 48.3 | | 140.88 | 122971.1 | 5028.6 76194.0 502.9 | 46.84 | 147599.7 0.0 100965.3 107664.6 | 8.11 8.05 -0.99 | 5.94 2257.5 1899.0 1220.7 | 32.73 | 0.60 67219.5 75835.1 101602.6 48.0 | | 141.88 | 122468.3 | 5531.4 75691.1 502.9 | 46.71 | 147599.7 0.0 101208.3 107436.3 | 8.16 8.10 -0.97 | 5.94 2263.8 1906.7 1220.3 | 32.62 | 0.62 69099.8 77055.5 103778.5 47.6 | | 142.88 | 121965.4 | 6034.3 75188.3 502.9 | 46.58 | 147599.7 0.0 101451.9 107206.2 | 8.21 8.16 -0.95 | 5.94 2270.1 1914.5 1219.9 | 32.50 | 0.64 70987.4 78275.6 105962.6 47.3 | | 143.88 | 121462.5 | 6537.2 74685.4 502.9 | 46.45 | 147599.7 0.0 101696.2 106974.5 | 8.26 8.21 -0.93 | 5.94 2276.5 1922.3 1219.5 | 32.39 | 0.65 72882.3 79495.2 108154.9 47.0 | | 144.88 | 120959.7 | 7040.0 74182.6 502.9 | 46.32 | 147599.7 0.0 101941.1 106741.2 | 8.31 8.26 -0.91 | 5.94 2283.0 1930.2 1219.1 | 32.28 | 0.67 74784.7 80714.6 110355.3 46.7 | | 145.88 | 120456.8 | 7542.9 73679.7 502.9 | 46.19 | 147599.7 0.0 102186.7 106506.1 | 8.37 8.32 -0.89 | 5.94 2289.5 1938.1 1218.8 | 32.16 | 0.69 76694.5 81933.5 112563.9 46.4 | | 146.88 | 119954.0 | 8045.7 73176.8 502.9 | 46.05 | 147599.7 0.0 102432.9 106269.3 | 8.42 8.37 -0.87 | 5.94 2296.1 1946.1 1218.5 | 32.05 | 0.71 78611.8 83152.2 114780.6 46.1 | | 147.88 | 119451.1 | 8548.6 72674.0 502.9 | 45.92 | 147599.7 0.0 102679.7 106030.9 | 8.47 8.43 -0.85 | 5.94 2302.8 1954.1 1218.3 | 31.94 | 0.72 80536.6 84370.6 117005.3 45.8 | | 148.88 | 118948.3 | 9051.4 72171.1 502.9 | 45.79 | 147599.7 0.0 102927.2 105790.6 | 8.53 8.49 -0.83 | 5.94 2309.5 1962.2 1218.0 | 31.83 | 0.74 82469.0 85588.7 119238.1 45.5 | | 149.88 | 118445.4 | 9554.3 71668.3 502.9 | 45.65 | 147599.7 0.0 103175.3 105548.7 | 8.58 8.54 -0.81 | 5.94 2316.3 1970.4 1217.8 | 31.72 | 0.76 84409.1 86806.6 121479.0 45.2 | | 150.88 | 117942.5 | 10057.2 71165.4 502.9 | 45.52 | 147599.7 0.0 103424.0 105305.0 | 8.64 8.60 -0.79 | 5.94 2323.2 1978.6 1217.6 | 31.61 | 0.78 86356.8 88024.3 123727.9 45.0 | | 151.88 | 117439.7 | 10560.0 70662.6 502.9 | 45.38 | 147599.7 0.0 103673.4 105059.5 | 8.69 8.66 -0.77 | 5.94 2330.1 1986.8 1217.4 | 31.50 | 0.79 88312.3 89241.8 125984.8 44.7 | | 152.88 | 116936.8 | 11062.9 70159.7 502.9 | 45.24 | 147599.7 0.0 103923.3 104812.3 | 8.75 8.72 -0.74 | 5.94 2337.2 1995.1 1217.3 | 31.39 | 0.81 90275.6 90459.2 128249.8 44.4 | | 153.88 | 116434.0 | 11565.7 69656.8 502.9 | 45.11 | 147599.7 0.0 104173.9 104563.2 | 8.80 8.77 -0.72 | 5.94 2344.2 2003.5 1217.2 | 31.28 | 0.83 92246.8 91676.4 130522.8 44.2 | | 154.88 | 115931.1 | 12068.6 69154.0 502.9 | 44.97 | 147599.7 0.0 104425.1 104312.4 | 8.86 8.83 -0.70 | 5.94 2351.4 2011.9 1217.1 | 31.17 | 0.85 94225.8 92893.6 132803.9 44.0 | | 155.88 | 115428.3 | 12571.4 68651.1 502.9 | 44.83 | 147599.7 0.0 104676.9 104059.7 | 8.92 8.89 -0.68 | 5.94 2358.6 2020.4 1217.0 | 31.06 | 0.86 96212.8 94110.6 135093.0 43.7 | | 156.88 | 114925.4 | 13074.3 68148.3 502.9 | 44.69 | 147599.7 0.0 104929.3 103805.2 | 8.98 8.95 -0.66 | 5.94 2365.9 2028.9 1217.0 | 30.96 | 0.88 98207.9 95327.6 137390.3 43.5 | | 157.88 | 114422.5 | 13577.2 67645.4 502.9 | 44.55 | 147599.7 0.0 105182.2 103548.9 | 9.04 9.01 -0.64 | 5.94 2373.3 2037.5 1217.0 | 30.85 | 0.90 100211.0 96544.6 139695.7 43.3 | | 158.88 | 113919.7 | 14080.0 67142.5 502.9 | 44.41 | 147599.7 0.0 105435.8 103290.7 | 9.10 9.08 -0.62 | 5.94 2380.8 2046.2 1217.0 | 30.74 | 0.92 102222.2 97761.6 142009.3 43.0 | | 159.88 | 113416.8 | 14582.9 66639.7 502.9 | 44.27 | 147599.7 0.0 105690.0 103030.6 | 9.16 9.14 -0.60 | 5.94 2388.3 2054.9 1217.0 | 30.64 | 0.94 104241.6 98978.5 144331.1 42.8 | | 160.88 | 112914.0 | 15085.7 66136.8 502.9 | 44.13 | 147599.7 0.0 105944.7 102768.6 | 9.22 9.20 -0.58 | 5.94 2395.9 2063.7 1217.1 | 30.53 | 0.95 106269.3 100195.6 146661.0 42.6 | | 161.88 | 112411.1 | 15588.6 65634.0 502.9 | 43.99 | 147599.7 0.0 106200.0 102504.8 | 9.28 9.26 -0.56 | 5.94 2403.5 2072.6 1217.2 | 30.42 | 0.97 108305.2 101412.7 148999.3 42.4 | | 162.88 | 111908.2 | 16091.5 65131.1 502.9 | 43.84 | 147599.7 0.0 106455.9 102239.0 | 9.34 9.33 -0.54 | 5.94 2411.3 2081.5 1217.3 | 30.32 | 0.99 110349.5 102630.0 151345.8 42.2 | | 163.88 | 111405.4 | 16594.3 64628.3 502.9 | 43.70 | 147599.7 0.0 106712.4 101971.3 | 9.41 9.39 -0.52 | 5.94 2419.1 2090.4 1217.4 | 30.22 | 1.01 112402.3 103847.3 153700.7 42.0 | | 164.88 | 110902.5 | 17097.2 64125.4 502.9 | 43.55 | 147599.7 0.0 106969.4 101701.6 | 9.47 9.46 -0.50 | 5.94 2427.0 2099.5 1217.6 | 30.11 | 1.03 114463.4 105064.8 156064.0 41.8 | | 165.88 | 110399.7 | 17600.0 63622.5 502.9 | 43.41 | 147599.7 0.0 107226.9 101430.1 | 9.54 9.52 -0.48 | 5.94 2435.0 2108.6 1217.8 | 30.01 | 1.05 116533.2 106282.5 158435.7 41.6 | | 166.88 | 109896.8 | 18102.9 63119.7 502.9 | 43.26 | 147599.7 0.0 107485.0 101156.5 | 9.60 9.59 -0.46 | 5.94 2443.0 2117.7 1218.0 | 29.91 | 1.07 118611.5 107500.4 160815.8 41.4 | | 167.88 | 109394.0 | 18605.7 62616.8 502.9 | 43.12 | 147599.7 0.0 107743.7 100881.0 | 9.67 9.66 -0.44 | 5.94 2451.1 2126.9 1218.3 | 29.80 | 1.08 120698.4 108718.6 163204.5 41.2 | | 168.88 | 108891.1 | 19108.6 62114.0 502.9 | 42.97 | 147599.7 0.0 108002.9 100603.4 | 9.74 9.73 -0.42 | 5.94 2459.3 2136.2 1218.5 | 29.70 | 1.10 122794.1 109937.0 165601.8 41.0 | Fairing Jettison: Time=168.93 sec, Altitude=110000 metres, 360893 feet | 169.88 | 107488.2 | 19611.5 61611.1 502.9 | 42.82 | 147599.7 0.0 108262.6 100323.9 | 9.88 9.88 -0.32 | 5.94 2467.7 2145.7 1218.9 | 29.60 | 1.12 124898.6 111155.7 168007.7 40.9 | | 170.88 | 106985.4 | 20114.3 61108.2 502.9 | 42.67 | 147599.7 0.0 108522.8 100042.3 | 9.95 9.95 -0.30 | 5.94 2476.2 2155.2 1219.3 | 29.50 | 1.14 127011.9 112374.8 170422.4 40.7 |

Ú | 285.88 | 49156.7 | 77943.0 3279.6 502.9 | 20.20 | 147599.7 0.0 138520.3 50968.6 | 27.66 27.63 1.12 | 5.94 4251.2 3996.4 1449.8 | 19.94 | 4.07 452928.6 262362.7 531344.8 27.5 | | 286.88 | 48653.9 | 78445.8 2776.7 502.9 | 19.96 | 147599.7 0.0 138735.6 50379.6 | 27.99 27.96 1.11 | 5.94 4277.7 4023.3 1453.3 | 19.86 | 4.10 456779.5 263814.2 535521.0 27.4 | | 287.88 | 48151.0 | 78948.7 2273.9 502.9 | 19.71 | 147599.7 0.0 138949.1 49787.8 | 28.32 28.30 1.10 | 5.94 4304.6 4050.5 1456.9 | 19.78 | 4.14 460655.6 265269.3 539723.1 27.3 | | 288.88 | 47648.1 | 79451.6 1771.0 502.9 | 19.47 | 147599.7 0.0 139160.6 49193.4 | 28.66 28.64 1.09 | 5.94 4331.7 4078.1 1460.5 | 19.70 | 4.17 464557.2 266728.0 543951.3 27.3 | | 289.88 | 47145.3 | 79954.4 1268.1 502.9 | 19.22 | 147599.7 0.0 139370.3 48596.2 | 29.01 28.99 1.08 | 5.94 4359.2 4106.0 1464.1 | 19.62 | 4.21 468484.6 268190.3 548205.9 27.2 | | 290.88 | 46642.4 | 80457.3 765.3 502.9 | 18.98 | 147599.7 0.0 139578.0 47996.3 | 29.37 29.35 1.06 | 5.94 4387.1 4134.3 1467.7 | 19.55 | 4.24 472438.0 269656.1 552487.3 27.1 | | 291.88 | 46139.6 | 80960.1 262.4 502.9 | 18.73 | 147599.7 0.0 139783.8 47393.8 | 29.73 29.71 1.05 | 5.94 4415.3 4162.9 1471.3 | 19.47 | 4.28 476417.9 271125.6 556795.7 27.0 | | 292.40 | 45877.1 | 81222.6 0.0 502.9 | 18.60 | 147599.7 0.0 139890.3 47078.3 | 29.92 29.90 1.04 | 5.94 4430.1 4178.0 1473.2 | 19.42 | 4.30 478505.5 271894.0 559055.0 26.9 | 182

Appendix B2 Sample TRAJ2DF Data Files

Traj2d - Fairing Coast to Impact

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | A Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 168.93 | 900.0 | -9.5 -9.5 | 5.94 2459.8 2136.72 1218.6 | 29.70 | 1.10 122903.3 110000.3 165726.7 41.0 | | 169.93 | 900.0 | -9.5 -9.5 | 5.94 2455.1 2136.32 1209.8 | 29.52 | 1.12 125003.4 111214.5 168126.3 40.8 | | 170.93 | 900.0 | -9.5 -9.5 | 5.94 2450.4 2135.93 1201.0 | 29.35 | 1.14 127102.7 112419.9 170521.4 40.7 | | 171.93 | 900.0 | -9.5 -9.5 | 5.94 2445.8 2135.53 1192.2 | 29.17 | 1.16 129201.3 113616.5 172911.7 40.5 | | 172.93 | 900.0 | -9.5 -9.5 | 5.94 2441.2 2135.14 1183.5 | 29.00 | 1.18 131299.0 114804.4 175297.4 40.3 | | 173.93 | 900.0 | -9.5 -9.5 | 5.94 2436.6 2134.76 1174.7 | 28.82 | 1.20 133396.0 115983.5 177678.3 40.1 | | 174.93 | 900.0 | -9.5 -9.5 | 5.94 2432.1 2134.37 1166.0 | 28.65 | 1.22 135492.3 117153.8 180054.7 40.0 | | 175.93 | 900.0 | -9.5 -9.5 | 5.94 2427.6 2133.99 1157.2 | 28.47 | 1.24 137587.8 118315.4 182426.3 39.8 | | 176.93 | 900.0 | -9.4 -9.4 | 5.94 2423.1 2133.61 1148.5 | 28.29 | 1.25 139682.5 119468.2 184793.2 39.6 | | 177.93 | 900.0 | -9.4 -9.4 | 5.94 2418.6 2133.23 1139.7 | 28.11 | 1.27 141776.6 120612.3 187155.5 39.5 | | 178.93 | 900.0 | -9.4 -9.4 | 5.94 2414.2 2132.86 1131.0 | 27.94 | 1.29 143869.8 121747.7 189513.1 39.3 | | 179.93 | 900.0 | -9.4 -9.4 | 5.94 2409.8 2132.49 1122.2 | 27.76 | 1.31 145962.4 122874.3 191866.1 39.2 | | 180.93 | 900.0 | -9.4 -9.4 | 5.94 2405.4 2132.13 1113.5 | 27.58 | 1.33 148054.2 123992.1 194214.4 39.0 | | 181.93 | 900.0 | -9.4 -9.4 | 5.94 2401.0 2131.76 1104.8 | 27.39 | 1.35 150145.3 125101.3 196558.0 38.9 | | 182.93 | 900.0 | -9.4 -9.4 | 5.94 2396.7 2131.40 1096.0 | 27.21 | 1.37 152235.7 126201.7 198897.0 38.7 | | 183.93 | 900.0 | -9.4 -9.4 | 5.94 2392.4 2131.04 1087.3 | 27.03 | 1.39 154325.4 127293.3 201231.4 38.5 | | 184.93 | 900.0 | -9.4 -9.4 | 5.94 2388.1 2130.69 1078.6 | 26.85 | 1.41 156414.4 128376.2 203561.1 38.4 | | 185.93 | 900.0 | -9.4 -9.4 | 5.94 2383.9 2130.34 1069.8 | 26.67 | 1.42 158502.7 129450.4 205886.2 38.2 | | 186.93 | 900.0 | -9.4 -9.4 | 5.94 2379.7 2129.99 1061.1 | 26.48 | 1.44 160590.3 130515.9 208206.6 38.1 | | 187.93 | 900.0 | -9.4 -9.4 | 5.94 2375.5 2129.64 1052.4 | 26.30 | 1.46 162677.3 131572.7 210522.5 37.9 | | 188.93 | 900.0 | -9.4 -9.4 | 5.94 2371.3 2129.30 1043.7 | 26.11 | 1.48 164763.5 132620.7 212833.8 37.8 | | 189.93 | 900.0 | -9.4 -9.4 | 5.94 2367.2 2128.96 1035.0 | 25.93 | 1.50 166849.1 133660.1 215140.4 37.7 | | 190.93 | 900.0 | -9.4 -9.4 | 5.94 2363.1 2128.62 1026.3 | 25.74 | 1.52 168934.1 134690.7 217442.5 37.5 | | 191.93 | 900.0 | -9.4 -9.4 | 5.94 2359.0 2128.29 1017.6 | 25.55 | 1.54 171018.3 135712.6 219740.1 37.4 | | 192.93 | 900.0 | -9.4 -9.4 | 5.94 2355.0 2127.96 1008.8 | 25.37 | 1.56 173102.0 136725.8 222033.1 37.2 | | 193.93 | 900.0 | -9.4 -9.4 | 5.94 2351.0 2127.63 1000.1 | 25.18 | 1.57 175184.9 137730.3 224321.5 37.1 | | 194.93 | 900.0 | -9.4 -9.4 | 5.94 2347.0 2127.31 991.4 | 24.99 | 1.59 177267.3 138726.1 226605.4 36.9 |

Ú | 501.93 | 900.0 | -9.7 -9.7 | 5.94 2728.8 2160.97 -1666.2 | -37.63 | 7.31 813214.9 37194.1 815878.5 -1.0 | | 502.93 | 900.0 | -9.7 -9.7 | 5.94 2734.7 2161.54 -1675.2 | -37.78 | 7.32 815363.9 35523.4 817846.6 -1.2 | | 503.93 | 900.0 | -9.7 -9.7 | 5.94 2740.6 2162.10 -1684.2 | -37.92 | 7.34 817514.0 33843.7 819818.0 -1.3 | | 504.93 | 900.0 | -9.7 -9.7 | 5.94 2746.6 2162.67 -1693.1 | -38.06 | 7.36 819665.3 32155.0 821792.8 -1.4 | | 505.93 | 900.0 | -9.7 -9.7 | 5.94 2752.6 2163.25 -1702.1 | -38.20 | 7.38 821817.7 30457.4 823771.1 -1.6 | | 506.93 | 900.0 | -9.7 -9.7 | 5.94 2758.6 2163.82 -1711.1 | -38.34 | 7.40 823971.2 28750.8 825752.8 -1.7 | | 507.93 | 900.0 | -9.7 -9.7 | 5.94 2764.7 2164.40 -1720.1 | -38.47 | 7.42 826125.9 27035.2 827738.0 -1.8 | | 508.93 | 900.0 | -9.7 -9.7 | 5.94 2770.7 2164.98 -1729.1 | -38.61 | 7.44 828281.7 25310.6 829726.7 -2.0 | | 509.93 | 900.0 | -9.7 -9.7 | 5.94 2776.8 2165.57 -1738.1 | -38.75 | 7.46 830438.8 23577.1 831718.9 -2.1 | | 510.93 | 900.0 | -9.7 -9.7 | 5.94 2782.9 2166.16 -1747.1 | -38.89 | 7.48 832596.9 21834.5 833714.8 -2.2 | | 511.93 | 900.0 | -9.7 -9.7 | 5.94 2789.0 2166.75 -1756.1 | -39.02 | 7.50 834756.3 20082.9 835714.2 -2.4 | | 512.93 | 900.0 | -9.8 -9.8 | 5.94 2795.2 2167.35 -1765.1 | -39.16 | 7.52 836916.8 18322.3 837717.3 -2.5 | | 513.93 | 900.0 | -9.8 -9.8 | 5.94 2801.3 2167.95 -1774.1 | -39.29 | 7.54 839078.6 16552.7 839724.1 -2.6 | | 514.93 | 900.0 | -9.8 -9.8 | 5.94 2807.5 2168.55 -1783.2 | -39.43 | 7.56 841241.5 14774.0 841734.6 -2.8 | | 515.93 | 900.0 | -9.8 -9.8 | 5.94 2813.7 2169.16 -1792.2 | -39.56 | 7.58 843405.7 12986.4 843748.8 -2.9 | | 516.93 | 900.0 | -9.8 -9.8 | 5.94 2820.0 2169.77 -1801.2 | -39.70 | 7.60 845571.0 11189.7 845766.8 -3.0 | | 517.93 | 900.0 | -9.8 -9.8 | 5.94 2826.2 2170.38 -1810.3 | -39.83 | 7.62 847737.6 9383.9 847788.6 -3.2 | | 518.93 | 900.0 | -9.8 -9.8 | 5.94 2832.5 2171.00 -1819.3 | -39.96 | 7.63 849905.4 7569.2 849814.2 -3.3 | | 519.93 | 900.0 | -9.8 -9.8 | 5.94 2838.8 2171.62 -1828.3 | -40.09 | 7.65 852074.5 5745.3 851843.7 -3.4 | | 520.93 | 900.0 | -9.8 -9.8 | 5.94 2845.1 2172.24 -1837.4 | -40.23 | 7.67 854244.7 3912.5 853877.1 -3.6 | | 521.93 | 900.0 | -9.8 -9.8 | 5.94 2851.4 2172.87 -1846.5 | -40.36 | 7.69 856416.3 2070.6 855914.5 -3.7 | | 522.93 | 900.0 | -9.8 -9.8 | 5.94 2857.8 2173.50 -1855.5 | -40.49 | 7.71 858589.1 219.6 857955.8 -3.8 | | 523.05 | 900.0 | -9.8 -9.8 | 5.94 2858.6 2173.58 -1856.6 | -40.50 | 7.72 858846.2 0.0 858197.5 -3.9 | 183

Appendix B2 Sample TRAJ2DF Data Files

Traj2d - Second Stage Coast to Impact

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | A Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 292.40 | 24113.6 | -9.0 -9.0 | 5.94 4430.1 4177.97 1473.2 | 19.42 | 4.30 478505.5 271894.0 559055.0 26.9 | | 293.40 | 24113.6 | -9.0 -9.0 | 5.94 4427.1 4177.05 1466.8 | 19.35 | 4.33 482511.7 273364.0 563389.9 26.8 | | 294.40 | 24113.6 | -9.0 -9.0 | 5.94 4424.1 4176.13 1460.5 | 19.28 | 4.37 486516.2 274827.7 567721.5 26.7 | | 295.40 | 24113.6 | -9.0 -9.0 | 5.94 4421.2 4175.21 1454.1 | 19.20 | 4.41 490519.0 276285.0 572049.9 26.7 | | 296.40 | 24113.6 | -9.0 -9.0 | 5.94 4418.2 4174.30 1447.7 | 19.13 | 4.44 494519.9 277735.8 576375.1 26.6 | | 297.40 | 24113.6 | -9.0 -9.0 | 5.94 4415.3 4173.40 1441.3 | 19.05 | 4.48 498519.2 279180.3 580697.1 26.5 | | 298.40 | 24113.6 | -9.0 -9.0 | 5.94 4412.3 4172.50 1434.9 | 18.98 | 4.51 502516.7 280618.4 585015.9 26.4 | | 299.40 | 24113.6 | -9.0 -9.0 | 5.94 4409.4 4171.60 1428.5 | 18.90 | 4.55 506512.5 282050.1 589331.5 26.3 | | 300.40 | 24113.6 | -9.0 -9.0 | 5.94 4406.5 4170.71 1422.1 | 18.83 | 4.59 510506.6 283475.5 593643.9 26.2 | | 301.40 | 24113.6 | -9.0 -9.0 | 5.94 4403.6 4169.82 1415.8 | 18.75 | 4.62 514499.0 284894.4 597953.1 26.1 | | 302.40 | 24113.6 | -9.0 -9.0 | 5.94 4400.7 4168.94 1409.4 | 18.68 | 4.66 518489.6 286307.0 602259.1 26.0 | | 303.40 | 24113.6 | -9.0 -9.0 | 5.94 4397.9 4168.06 1403.0 | 18.60 | 4.69 522478.6 287713.2 606562.0 25.9 | | 304.40 | 24113.6 | -9.0 -9.0 | 5.94 4395.0 4167.18 1396.6 | 18.53 | 4.73 526466.0 289113.0 610861.6 25.9 | | 305.40 | 24113.6 | -9.0 -9.0 | 5.94 4392.2 4166.31 1390.3 | 18.45 | 4.77 530451.6 290506.5 615158.0 25.8 | | 306.40 | 24113.6 | -9.0 -9.0 | 5.94 4389.3 4165.44 1383.9 | 18.38 | 4.80 534435.6 291893.6 619451.3 25.7 | | 307.40 | 24113.6 | -9.0 -9.0 | 5.94 4386.5 4164.58 1377.5 | 18.30 | 4.84 538417.9 293274.3 623741.4 25.6 | | 308.40 | 24113.6 | -9.0 -9.0 | 5.94 4383.7 4163.72 1371.2 | 18.23 | 4.87 542398.6 294648.7 628028.3 25.5 | | 309.40 | 24113.6 | -9.0 -9.0 | 5.94 4380.9 4162.87 1364.8 | 18.15 | 4.91 546377.7 296016.7 632312.0 25.4 | | 310.40 | 24113.6 | -9.0 -9.0 | 5.94 4378.1 4162.02 1358.5 | 18.08 | 4.94 550355.1 297378.3 636592.6 25.3 | | 311.40 | 24113.6 | -8.9 -8.9 | 5.94 4375.3 4161.18 1352.1 | 18.00 | 4.98 554330.9 298733.6 640870.0 25.3 | | 312.40 | 24113.6 | -8.9 -8.9 | 5.94 4372.6 4160.34 1345.8 | 17.92 | 5.02 558305.1 300082.5 645144.2 25.2 | | 313.40 | 24113.6 | -8.9 -8.9 | 5.94 4369.8 4159.50 1339.4 | 17.85 | 5.05 562277.7 301425.1 649415.2 25.1 | | 314.40 | 24113.6 | -8.9 -8.9 | 5.94 4367.1 4158.67 1333.1 | 17.77 | 5.09 566248.7 302761.3 653683.2 25.0 | | 315.40 | 24113.6 | -8.9 -8.9 | 5.94 4364.4 4157.84 1326.7 | 17.70 | 5.12 570218.1 304091.2 657947.9 24.9 | | 316.40 | 24113.6 | -8.9 -8.9 | 5.94 4361.7 4157.02 1320.4 | 17.62 | 5.16 574185.9 305414.8 662209.5 24.9 | | 317.40 | 24113.6 | -8.9 -8.9 | 5.94 4359.0 4156.20 1314.0 | 17.54 | 5.19 578152.2 306731.9 666468.0 24.8 | | 318.40 | 24113.6 | -8.9 -8.9 | 5.94 4356.3 4155.38 1307.7 | 17.47 | 5.23 582116.9 308042.8 670723.3 24.7 |

Ú | 885.40 | 24113.6 | -9.7 -9.7 | 5.94 4876.6 4322.66 -2257.3 | -27.57 | 25.21 2806193.0 49306.8 2794787.8 -11.6 | | 886.40 | 24113.6 | -9.7 -9.7 | 5.94 4881.0 4324.18 -2264.1 | -27.64 | 25.25 2810484.0 47046.1 2798459.7 -11.7 | | 887.40 | 24113.6 | -9.7 -9.7 | 5.94 4885.5 4325.71 -2270.8 | -27.70 | 25.29 2814778.0 44778.6 2802133.0 -11.7 | | 888.40 | 24113.6 | -9.7 -9.7 | 5.94 4890.0 4327.24 -2277.6 | -27.76 | 25.32 2819075.1 42504.4 2805807.8 -11.8 | | 889.40 | 24113.6 | -9.7 -9.7 | 5.94 4894.6 4328.78 -2284.4 | -27.82 | 25.36 2823375.2 40223.4 2809484.0 -11.9 | | 890.40 | 24113.6 | -9.7 -9.7 | 5.94 4899.1 4330.32 -2291.1 | -27.88 | 25.40 2827678.4 37935.7 2813161.7 -11.9 | | 891.40 | 24113.6 | -9.7 -9.7 | 5.94 4903.6 4331.87 -2297.9 | -27.94 | 25.44 2831984.6 35641.2 2816840.8 -12.0 | | 892.40 | 24113.6 | -9.7 -9.7 | 5.94 4908.2 4333.42 -2304.7 | -28.01 | 25.48 2836294.0 33339.9 2820521.4 -12.1 | | 893.40 | 24113.6 | -9.7 -9.7 | 5.94 4912.7 4334.98 -2311.5 | -28.07 | 25.52 2840606.4 31031.8 2824203.5 -12.1 | | 894.40 | 24113.6 | -9.7 -9.7 | 5.94 4917.3 4336.55 -2318.2 | -28.13 | 25.56 2844922.0 28717.0 2827887.1 -12.2 | | 895.40 | 24113.6 | -9.7 -9.7 | 5.94 4921.9 4338.12 -2325.0 | -28.19 | 25.60 2849240.6 26395.3 2831572.1 -12.3 | | 896.40 | 24113.6 | -9.7 -9.7 | 5.94 4926.5 4339.70 -2331.8 | -28.25 | 25.63 2853562.5 24066.9 2835258.8 -12.3 | | 897.40 | 24113.6 | -9.7 -9.7 | 5.94 4931.1 4341.28 -2338.6 | -28.31 | 25.67 2857887.4 21731.7 2838946.9 -12.4 | | 898.40 | 24113.6 | -9.7 -9.7 | 5.94 4935.7 4342.87 -2345.4 | -28.37 | 25.71 2862215.5 19389.7 2842636.6 -12.5 | | 899.40 | 24113.6 | -9.8 -9.8 | 5.94 4940.4 4344.47 -2352.2 | -28.43 | 25.75 2866546.8 17040.9 2846327.8 -12.5 | | 900.40 | 24113.6 | -9.8 -9.8 | 5.94 4945.0 4346.07 -2359.0 | -28.49 | 25.79 2870881.3 14685.3 2850020.6 -12.6 | | 901.40 | 24113.6 | -9.8 -9.8 | 5.94 4949.7 4347.68 -2365.8 | -28.55 | 25.83 2875219.0 12322.9 2853714.9 -12.7 | | 902.40 | 24113.6 | -9.8 -9.8 | 5.94 4954.4 4349.29 -2372.6 | -28.61 | 25.87 2879559.9 9953.7 2857410.8 -12.7 | | 903.40 | 24113.6 | -9.8 -9.8 | 5.94 4959.0 4350.91 -2379.4 | -28.67 | 25.91 2883904.0 7577.6 2861108.3 -12.8 | | 904.40 | 24113.6 | -9.8 -9.8 | 5.94 4963.7 4352.53 -2386.3 | -28.73 | 25.95 2888251.4 5194.8 2864807.5 -12.9 | | 905.40 | 24113.6 | -9.8 -9.8 | 5.94 4968.5 4354.16 -2393.1 | -28.79 | 25.98 2892602.0 2805.1 2868508.2 -12.9 | | 906.40 | 24113.6 | -9.8 -9.8 | 5.94 4973.2 4355.80 -2399.9 | -28.85 | 26.02 2896955.9 408.6 2872210.5 -13.0 | | 906.57 | 24113.6 | -9.8 -9.8 | 5.94 4974.0 4356.08 -2401.1 | -28.86 | 26.03 2897697.3 0.0 2872840.8 -13.0 | 184

Appendix B2 Sample TRAJ2DF Data Files

Traj2d - Stage 2/3 Interstage Coast

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | A Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 292.40 | 45877.1 | -9.0 -9.0 | 5.94 4430.1 4177.97 1473.2 | 19.42 | 4.30 478505.5 271894.0 559055.0 26.9 | | 293.40 | 45877.1 | -9.0 -9.0 | 5.94 4427.1 4177.05 1466.8 | 19.35 | 4.33 482511.7 273364.0 563389.9 26.8 | | 294.40 | 45877.1 | -9.0 -9.0 | 5.94 4424.1 4176.13 1460.5 | 19.28 | 4.37 486516.2 274827.7 567721.5 26.7 | | 295.40 | 45877.1 | -9.0 -9.0 | 5.94 4421.2 4175.21 1454.1 | 19.20 | 4.41 490519.0 276285.0 572049.9 26.7 | | 296.40 | 45877.1 | -9.0 -9.0 | 5.94 4418.2 4174.30 1447.7 | 19.13 | 4.44 494519.9 277735.8 576375.1 26.6 | | 297.40 | 45877.1 | -9.0 -9.0 | 5.94 4415.3 4173.40 1441.3 | 19.05 | 4.48 498519.2 279180.3 580697.1 26.5 | | 298.40 | 45877.1 | -9.0 -9.0 | 5.94 4412.3 4172.50 1434.9 | 18.98 | 4.51 502516.7 280618.4 585015.9 26.4 | | 299.40 | 45877.1 | -9.0 -9.0 | 5.94 4409.4 4171.60 1428.5 | 18.90 | 4.55 506512.5 282050.1 589331.5 26.3 | | 300.40 | 45877.1 | -9.0 -9.0 | 5.94 4406.5 4170.71 1422.1 | 18.83 | 4.59 510506.6 283475.5 593643.9 26.2 | | 301.40 | 45877.1 | -9.0 -9.0 | 5.94 4403.6 4169.82 1415.8 | 18.75 | 4.62 514499.0 284894.4 597953.1 26.1 | | 302.40 | 45877.1 | -9.0 -9.0 | 5.94 4400.7 4168.94 1409.4 | 18.68 | 4.66 518489.6 286307.0 602259.1 26.0 | | 303.40 | 45877.1 | -9.0 -9.0 | 5.94 4397.9 4168.06 1403.0 | 18.60 | 4.69 522478.6 287713.2 606562.0 25.9 | | 304.40 | 45877.1 | -9.0 -9.0 | 5.94 4395.0 4167.18 1396.6 | 18.53 | 4.73 526466.0 289113.0 610861.6 25.9 | | 305.40 | 45877.1 | -9.0 -9.0 | 5.94 4392.2 4166.31 1390.3 | 18.45 | 4.77 530451.6 290506.5 615158.0 25.8 | | 306.40 | 45877.1 | -9.0 -9.0 | 5.94 4389.3 4165.44 1383.9 | 18.38 | 4.80 534435.6 291893.6 619451.3 25.7 | | 307.40 | 45877.1 | -9.0 -9.0 | 5.94 4386.5 4164.58 1377.5 | 18.30 | 4.84 538417.9 293274.3 623741.4 25.6 | | 308.40 | 45877.1 | -9.0 -9.0 | 5.94 4383.7 4163.72 1371.2 | 18.23 | 4.87 542398.6 294648.7 628028.3 25.5 | | 309.40 | 45877.1 | -9.0 -9.0 | 5.94 4380.9 4162.87 1364.8 | 18.15 | 4.91 546377.7 296016.7 632312.0 25.4 | | 310.40 | 45877.1 | -9.0 -9.0 | 5.94 4378.1 4162.02 1358.5 | 18.08 | 4.94 550355.1 297378.3 636592.6 25.3 | | 311.40 | 45877.1 | -8.9 -8.9 | 5.94 4375.3 4161.18 1352.1 | 18.00 | 4.98 554330.9 298733.6 640870.0 25.3 | | 312.40 | 45877.1 | -8.9 -8.9 | 5.94 4372.6 4160.34 1345.8 | 17.92 | 5.02 558305.1 300082.5 645144.2 25.2 | | 313.40 | 45877.1 | -8.9 -8.9 | 5.94 4369.8 4159.50 1339.4 | 17.85 | 5.05 562277.7 301425.1 649415.2 25.1 | | 314.40 | 45877.1 | -8.9 -8.9 | 5.94 4367.1 4158.67 1333.1 | 17.77 | 5.09 566248.7 302761.3 653683.2 25.0 | | 315.40 | 45877.1 | -8.9 -8.9 | 5.94 4364.4 4157.84 1326.7 | 17.70 | 5.12 570218.1 304091.2 657947.9 24.9 | | 316.40 | 45877.1 | -8.9 -8.9 | 5.94 4361.7 4157.02 1320.4 | 17.62 | 5.16 574185.9 305414.8 662209.5 24.9 | | 317.40 | 45877.1 | -8.9 -8.9 | 5.94 4359.0 4156.20 1314.0 | 17.54 | 5.19 578152.2 306731.9 666468.0 24.8 | | 318.40 | 45877.1 | -8.9 -8.9 | 5.94 4356.3 4155.38 1307.7 | 17.47 | 5.23 582116.9 308042.8 670723.3 24.7 | | 319.40 | 45877.1 | -8.9 -8.9 | 5.94 4353.6 4154.57 1301.3 | 17.39 | 5.26 586080.1 309347.3 674975.5 24.6 | | 320.40 | 45877.1 | -8.9 -8.9 | 5.94 4351.0 4153.77 1295.0 | 17.32 | 5.30 590041.7 310645.4 679224.6 24.5 | | 321.40 | 45877.1 | -8.9 -8.9 | 5.94 4348.3 4152.96 1288.7 | 17.24 | 5.34 594001.8 311937.3 683470.5 24.5 | | 322.40 | 45877.1 | -8.9 -8.9 | 5.94 4345.7 4152.17 1282.3 | 17.16 | 5.37 597960.3 313222.7 687713.3 24.4 | | 323.40 | 45877.1 | -8.9 -8.9 | 5.94 4343.0 4151.37 1276.0 | 17.09 | 5.41 601917.4 314501.9 691953.0 24.3 | | 324.40 | 45877.1 | -8.9 -8.9 | 5.94 4340.4 4150.58 1269.7 | 17.01 | 5.44 605872.9 315774.7 696189.6 24.2 | | 325.40 | 45877.1 | -8.9 -8.9 | 5.94 4337.8 4149.80 1263.3 | 16.93 | 5.48 609827.0 317041.2 700423.0 24.1 | | 326.40 | 45877.1 | -8.9 -8.9 | 5.94 4335.3 4149.02 1257.0 | 16.85 | 5.51 613779.5 318301.4 704653.4 24.1 | | 327.40 | 45877.1 | -8.9 -8.9 | 5.94 4332.7 4148.24 1250.7 | 16.78 | 5.55 617730.6 319555.2 708880.7 24.0 | | 327.80 | 45877.1 | -8.9 -8.9 | 5.94 4331.7 4147.93 1248.1 | 16.75 | 5.56 619310.6 320055.0 710570.7 24.0 |

185

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Third Stage Powered Flight

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Stage Propellant | Pitch | Thrust | Acceleration | Velocity | FPA | Distance from Pad | Orbit Elements | | | Mass | Used Remain Flow | | Total Tx Ty | A Ax Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | SMA Hp Ra Eccen ArgPg TruAn | | sec | kg | kg kg kg/s | deg | Kgf Kgf Kgf | m/s2 m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | km km km deg deg | | 327.80 | 36089.0 | 0.0 26727.3 33.7 | 7.80 | 15000.0 14861.2 2035.7 | 9.27 4.04 -8.34 | 5.94 4331.7 4147.9 1248.1 | 16.75 | 5.56 619310.6 320055.0 710570.7 24.0 | 3975.9 -5249.7 6823.4 0.716 263.0 187.0 | | 328.80 | 36055.3 | 33.7 26693.6 33.7 | 7.79 | 15000.0 14861.5 2033.3 | 9.26 4.04 -8.34 | 5.94 4333.1 4151.2 1242.4 | 16.66 | 5.60 623261.5 321300.3 714795.6 23.9 | 3977.3 -5247.1 6823.6 0.716 263.0 187.0 | | 329.80 | 36021.6 | 67.4 26659.9 33.7 | 7.78 | 15000.0 14861.9 2030.8 | 9.26 4.05 -8.33 | 5.94 4334.6 4154.5 1236.6 | 16.58 | 5.63 627214.8 322539.8 719021.1 23.8 | 3978.7 -5244.6 6823.8 0.715 263.1 186.9 | | 330.80 | 35987.9 | 101.1 26626.2 33.7 | 7.77 | 15000.0 14862.2 2028.4 | 9.26 4.05 -8.33 | 5.94 4336.1 4157.8 1230.9 | 16.49 | 5.67 631170.5 323773.5 723247.4 23.7 | 3980.1 -5242.0 6824.0 0.715 263.1 186.9 | | 331.80 | 35954.2 | 134.8 26592.5 33.7 | 7.76 | 15000.0 14862.5 2025.9 | 9.26 4.05 -8.33 | 5.94 4337.7 4161.0 1225.1 | 16.41 | 5.71 635128.6 325001.5 727474.4 23.6 | 3981.5 -5239.4 6824.2 0.714 263.1 186.9 | | 332.80 | 35920.5 | 168.5 26558.8 33.7 | 7.75 | 15000.0 14862.9 2023.5 | 9.26 4.06 -8.32 | 5.94 4339.2 4164.3 1219.4 | 16.32 | 5.74 639089.1 326223.7 731702.1 23.6 | 3982.9 -5236.8 6824.4 0.713 263.1 186.9 | | 333.80 | 35886.8 | 202.2 26525.1 33.7 | 7.74 | 15000.0 14863.2 2021.0 | 9.26 4.06 -8.32 | 5.94 4340.8 4167.6 1213.6 | 16.24 | 5.78 643052.0 327440.3 735930.7 23.5 | 3984.3 -5234.2 6824.6 0.713 263.1 186.9 | | 334.80 | 35853.1 | 235.9 26491.4 33.7 | 7.73 | 15000.0 14863.5 2018.5 | 9.26 4.07 -8.32 | 5.94 4342.3 4171.0 1207.9 | 16.15 | 5.81 647017.4 328651.0 740160.0 23.4 | 3985.7 -5231.6 6824.8 0.712 263.2 186.8 | | 335.80 | 35819.5 | 269.5 26457.7 33.7 | 7.72 | 15000.0 14863.9 2016.1 | 9.26 4.07 -8.31 | 5.94 4343.9 4174.3 1202.2 | 16.07 | 5.85 650985.2 329856.1 744390.1 23.3 | 3987.1 -5229.0 6825.0 0.712 263.2 186.8 | | 336.80 | 35785.8 | 303.2 26424.0 33.7 | 7.71 | 15000.0 14864.2 2013.6 | 9.26 4.07 -8.31 | 5.94 4345.6 4177.6 1196.5 | 15.98 | 5.88 654955.4 331055.4 748620.9 23.3 | 3988.5 -5226.4 6825.2 0.711 263.2 186.8 | | 337.80 | 35752.1 | 336.9 26390.3 33.7 | 7.71 | 15000.0 14864.5 2011.2 | 9.25 4.08 -8.31 | 5.94 4347.2 4180.9 1190.8 | 15.90 | 5.92 658928.1 332249.1 752852.7 23.2 | 3989.9 -5223.8 6825.4 0.711 263.2 186.8 | | 338.80 | 35718.4 | 370.6 26356.6 33.7 | 7.70 | 15000.0 14864.9 2008.7 | 9.25 4.08 -8.30 | 5.94 4348.9 4184.3 1185.1 | 15.81 | 5.95 662903.3 333437.0 757085.2 23.1 | 3991.3 -5221.2 6825.6 0.710 263.2 186.8 | | 339.80 | 35684.7 | 404.3 26322.9 33.7 | 7.69 | 15000.0 14865.2 2006.3 | 9.25 4.09 -8.30 | 5.94 4350.5 4187.6 1179.4 | 15.73 | 5.99 666880.9 334619.2 761318.6 23.0 | 3992.7 -5218.5 6825.8 0.710 263.3 186.7 | | 340.80 | 35651.0 | 438.0 26289.3 33.7 | 7.68 | 15000.0 14865.5 2003.8 | 9.25 4.09 -8.30 | 5.94 4352.2 4191.0 1173.7 | 15.65 | 6.03 670861.0 335795.8 765552.8 23.0 | 3994.1 -5215.9 6826.0 0.709 263.3 186.7 | | 341.80 | 35617.3 | 471.7 26255.6 33.7 | 7.67 | 15000.0 14865.9 2001.3 | 9.25 4.09 -8.30 | 5.94 4353.9 4194.3 1168.0 | 15.56 | 6.06 674843.6 336966.7 769788.0 22.9 | 3995.6 -5213.2 6826.2 0.708 263.3 186.7 | | 342.80 | 35583.6 | 505.4 26221.9 33.7 | 7.66 | 15000.0 14866.2 1998.9 | 9.25 4.10 -8.29 | 5.94 4355.7 4197.7 1162.4 | 15.48 | 6.10 678828.7 338131.8 774024.0 22.8 | 3997.0 -5210.6 6826.4 0.708 263.3 186.7 | | 343.80 | 35549.9 | 539.1 26188.2 33.7 | 7.65 | 15000.0 14866.5 1996.4 | 9.25 4.10 -8.29 | 5.94 4357.4 4201.1 1156.7 | 15.39 | 6.13 682816.4 339291.4 778260.9 22.7 | 3998.4 -5207.9 6826.6 0.707 263.4 186.6 | | 344.80 | 35516.2 | 572.8 26154.5 33.7 | 7.64 | 15000.0 14866.8 1994.0 | 9.25 4.10 -8.29 | 5.94 4359.2 4204.5 1151.0 | 15.31 | 6.17 686806.5 340445.2 782498.7 22.7 | 3999.9 -5205.2 6826.8 0.707 263.4 186.6 | | 345.80 | 35482.5 | 606.5 26120.8 33.7 | 7.63 | 15000.0 14867.2 1991.5 | 9.25 4.11 -8.28 | 5.94 4360.9 4207.8 1145.4 | 15.23 | 6.21 690799.2 341593.4 786737.5 22.6 | 4001.3 -5202.5 6827.0 0.706 263.4 186.6 | | 346.80 | 35448.8 | 640.2 26087.1 33.7 | 7.62 | 15000.0 14867.5 1989.1 | 9.25 4.11 -8.28 | 5.94 4362.7 4211.2 1139.7 | 15.14 | 6.24 694794.4 342736.0 790977.1 22.5 | 4002.7 -5199.9 6827.2 0.706 263.4 186.6 | | 347.80 | 35415.1 | 673.9 26053.4 33.7 | 7.61 | 15000.0 14867.8 1986.6 | 9.25 4.12 -8.28 | 5.94 4364.6 4214.6 1134.1 | 15.06 | 6.28 698792.1 343872.9 795217.8 22.4 | 4004.2 -5197.2 6827.4 0.705 263.4 186.6 | | 348.80 | 35381.4 | 707.6 26019.7 33.7 | 7.60 | 15000.0 14868.2 1984.1 | 9.25 4.12 -8.28 | 5.94 4366.4 4218.0 1128.5 | 14.98 | 6.31 702792.5 345004.1 799459.4 22.4 | 4005.6 -5194.5 6827.6 0.704 263.5 186.5 | | 349.80 | 35347.7 | 741.3 25986.0 33.7 | 7.59 | 15000.0 14868.5 1981.7 | 9.24 4.13 -8.27 | 5.94 4368.2 4221.5 1122.8 | 14.89 | 6.35 706795.3 346129.8 803702.0 22.3 | 4007.1 -5191.7 6827.7 0.704 263.5 186.5 | | 350.80 | 35314.0 | 775.0 25952.3 33.7 | 7.58 | 15000.0 14868.8 1979.2 | 9.24 4.13 -8.27 | 5.94 4370.1 4224.9 1117.2 | 14.81 | 6.39 710800.8 347249.8 807945.6 22.2 | 4008.5 -5189.0 6827.9 0.703 263.5 186.5 | | 351.80 | 35280.4 | 808.6 25918.6 33.7 | 7.57 | 15000.0 14869.1 1976.8 | 9.24 4.13 -8.27 | 5.94 4372.0 4228.3 1111.6 | 14.73 | 6.42 714808.8 348364.2 812190.2 22.1 | 4010.0 -5186.3 6828.1 0.703 263.5 186.5 | | 352.80 | 35246.7 | 842.3 25884.9 33.7 | 7.56 | 15000.0 14869.5 1974.3 | 9.24 4.14 -8.26 | 5.94 4373.9 4231.8 1106.0 | 14.65 | 6.46 718819.5 349473.0 816435.8 22.1 | 4011.4 -5183.6 6828.3 0.702 263.5 186.5 | | 353.80 | 35213.0 | 876.0 25851.2 33.7 | 7.55 | 15000.0 14869.8 1971.8 | 9.24 4.14 -8.26 | 5.94 4375.8 4235.2 1100.4 | 14.56 | 6.49 722832.7 350576.2 820682.5 22.0 | 4012.9 -5180.8 6828.5 0.702 263.6 186.4 | | 354.80 | 35179.3 | 909.7 25817.5 33.7 | 7.54 | 15000.0 14870.1 1969.4 | 9.24 4.15 -8.26 | 5.94 4377.8 4238.7 1094.8 | 14.48 | 6.53 726848.6 351673.8 824930.2 21.9 | 4014.4 -5178.1 6828.7 0.701 263.6 186.4 | | 355.80 | 35145.6 | 943.4 25783.8 33.7 | 7.53 | 15000.0 14870.4 1966.9 | 9.24 4.15 -8.26 | 5.94 4379.7 4242.1 1089.2 | 14.40 | 6.57 730867.0 352765.8 829179.0 21.9 | 4015.8 -5175.3 6828.9 0.700 263.6 186.4 | | 356.80 | 35111.9 | 977.1 25750.2 33.7 | 7.53 | 15000.0 14870.8 1964.5 | 9.24 4.15 -8.25 | 5.94 4381.7 4245.6 1083.6 | 14.32 | 6.60 734888.1 353852.2 833428.8 21.8 | 4017.3 -5172.5 6829.0 0.700 263.6 186.4 | | 357.80 | 35078.2 | 1010.8 25716.5 33.7 | 7.52 | 15000.0 14871.1 1962.0 | 9.24 4.16 -8.25 | 5.94 4383.7 4249.1 1078.1 | 14.24 | 6.64 738911.9 354933.0 837679.7 21.7 | 4018.8 -5169.8 6829.2 0.699 263.6 186.4 | | 358.80 | 35044.5 | 1044.5 25682.8 33.7 | 7.51 | 15000.0 14871.4 1959.5 | 9.24 4.16 -8.25 | 5.94 4385.7 4252.5 1072.5 | 14.15 | 6.67 742938.3 356008.3 841931.8 21.6 | 4020.3 -5167.0 6829.4 0.699 263.7 186.3 |

Ú | 1099.80 | 10077.5 | 26011.5 715.8 33.7 | 0.40 | 15000.0 14999.6 106.0 | 17.22 14.60 -9.13 | 5.94 9911.4 9908.0 -259.4 | -1.50 | 47.23 5257161.2 195945.1 5191194.7 -21.6 | 17308.8 190.1 28049.4 0.621 86.1 3.9 | | 1100.80 | 10043.8 | 26045.2 682.1 33.7 | 0.40 | 15000.0 14999.6 103.5 | 17.26 14.65 -9.13 | 5.94 9926.2 9923.0 -253.6 | -1.46 | 47.31 5266781.4 195688.5 5200024.9 -21.7 | 17529.5 190.1 28490.7 0.625 86.2 3.8 | | 1101.80 | 10010.1 | 26078.9 648.4 33.7 | 0.39 | 15000.0 14999.6 100.9 | 17.30 14.69 -9.13 | 5.94 9941.1 9938.0 -247.7 | -1.43 | 47.40 5276416.5 195437.9 5208868.4 -21.7 | 17757.0 190.2 28945.7 0.630 86.3 3.7 | | 1102.80 | 9976.5 | 26112.5 614.7 33.7 | 0.38 | 15000.0 14999.6 98.4 | 17.35 14.74 -9.14 | 5.94 9956.1 9953.1 -241.8 | -1.39 | 47.49 5286066.5 195193.1 5217725.0 -21.8 | 17991.6 190.2 29414.9 0.635 86.4 3.6 | | 1103.80 | 9942.8 | 26146.2 581.0 33.7 | 0.37 | 15000.0 14999.7 95.9 | 17.39 14.79 -9.14 | 5.94 9971.1 9968.3 -235.9 | -1.36 | 47.57 5295731.6 194954.2 5226594.9 -21.8 | 18233.7 190.2 29899.1 0.640 86.5 3.5 | | 1104.80 | 9909.1 | 26179.9 547.3 33.7 | 0.36 | 15000.0 14999.7 93.4 | 17.43 14.84 -9.14 | 5.94 9986.1 9983.4 -229.9 | -1.32 | 47.66 5305411.8 194721.4 5235478.1 -21.9 | 18483.7 190.3 30399.0 0.645 86.6 3.4 | | 1105.80 | 9875.4 | 26213.6 513.6 33.7 | 0.35 | 15000.0 14999.7 90.8 | 17.48 14.90 -9.14 | 5.94 10001.2 9998.6 -223.8 | -1.28 | 47.75 5315107.0 194494.5 5244374.6 -21.9 | 18741.9 190.3 30915.3 0.650 86.7 3.3 | | 1106.80 | 9841.7 | 26247.3 479.9 33.7 | 0.34 | 15000.0 14999.7 88.3 | 17.52 14.95 -9.15 | 5.94 10016.310013.9 -217.7 | -1.25 | 47.83 5324817.3 194273.8 5253284.6 -22.0 | 19008.7 190.3 31449.0 0.654 86.9 3.1 | | 1107.80 | 9808.0 | 26281.0 446.3 33.7 | 0.33 | 15000.0 14999.7 85.8 | 17.57 15.00 -9.15 | 5.94 10031.410029.2 -211.6 | -1.21 | 47.92 5334542.8 194059.1 5262207.9 -22.0 | 19284.7 190.4 32000.9 0.659 87.0 3.0 | | 1108.80 | 9774.3 | 26314.7 412.6 33.7 | 0.32 | 15000.0 14999.7 83.3 | 17.61 15.05 -9.15 | 5.94 10046.610044.5 -205.4 | -1.17 | 48.01 5344283.5 193850.6 5271144.7 -22.1 | 19570.2 190.4 32571.9 0.664 87.1 2.9 | | 1109.80 | 9740.6 | 26348.4 378.9 33.7 | 0.31 | 15000.0 14999.8 80.7 | 17.66 15.10 -9.16 | 5.94 10061.910059.9 -199.2 | -1.13 | 48.10 5354039.3 193648.2 5280095.1 -22.1 | 19865.8 190.4 33163.0 0.669 87.2 2.8 | | 1110.80 | 9706.9 | 26382.1 345.2 33.7 | 0.30 | 15000.0 14999.8 78.2 | 17.71 15.15 -9.16 | 5.94 10077.210075.4 -192.9 | -1.10 | 48.18 5363810.5 193452.2 5289058.9 -22.2 | 20172.0 190.5 33775.4 0.674 87.3 2.7 | | 1111.80 | 9673.2 | 26415.8 311.5 33.7 | 0.29 | 15000.0 14999.8 75.7 | 17.75 15.21 -9.16 | 5.94 10092.610090.8 -186.6 | -1.06 | 48.27 5373596.9 193262.4 5298036.4 -22.2 | 20489.4 190.5 34410.2 0.679 87.4 2.6 | | 1112.80 | 9639.5 | 26449.5 277.8 33.7 | 0.28 | 15000.0 14999.8 73.2 | 17.80 15.26 -9.16 | 5.94 10108.010106.3 -180.3 | -1.02 | 48.36 5383398.6 193078.9 5307027.5 -22.3 | 20818.6 190.5 35068.6 0.684 87.5 2.5 | | 1113.80 | 9605.8 | 26483.2 244.1 33.7 | 0.27 | 15000.0 14999.8 70.6 | 17.85 15.31 -9.17 | 5.94 10123.410121.9 -173.9 | -0.98 | 48.45 5393215.7 192901.9 5316032.3 -22.3 | 21160.3 190.5 35752.0 0.690 87.6 2.4 | | 1114.80 | 9572.1 | 26516.9 210.4 33.7 | 0.26 | 15000.0 14999.8 68.1 | 17.90 15.37 -9.17 | 5.94 10138.910137.5 -167.4 | -0.95 | 48.54 5403048.2 192731.2 5325050.8 -22.4 | 21515.3 190.5 36461.9 0.695 87.7 2.3 | | 1115.80 | 9538.4 | 26550.6 176.7 33.7 | 0.25 | 15000.0 14999.8 65.6 | 17.94 15.42 -9.17 | 5.94 10154.410153.2 -160.9 | -0.91 | 48.62 5412896.0 192567.0 5334083.0 -22.4 | 21884.2 190.6 37199.8 0.700 87.8 2.2 | | 1116.80 | 9504.7 | 26584.3 143.0 33.7 | 0.24 | 15000.0 14999.8 63.1 | 17.99 15.48 -9.18 | 5.94 10170.010168.9 -154.4 | -0.87 | 48.71 5422759.4 192409.4 5343129.0 -22.5 | 22268.0 190.6 37967.4 0.705 87.9 2.1 | | 1117.80 | 9471.0 | 26618.0 109.3 33.7 | 0.23 | 15000.0 14999.8 60.5 | 18.04 15.53 -9.18 | 5.94 10185.710184.6 -147.8 | -0.83 | 48.80 5432638.2 192258.3 5352188.9 -22.5 | 22667.6 190.6 38766.5 0.710 88.0 2.0 | | 1118.80 | 9437.4 | 26651.6 75.6 33.7 | 0.22 | 15000.0 14999.9 58.0 | 18.09 15.59 -9.18 | 5.94 10201.310200.4 -141.2 | -0.79 | 48.89 5442532.5 192113.8 5361262.7 -22.6 | 23084.0 190.6 39599.2 0.715 88.1 1.9 | | 1119.80 | 9403.7 | 26685.3 41.9 33.7 | 0.21 | 15000.0 14999.9 55.5 | 18.14 15.64 -9.18 | 5.94 10217.110216.2 -134.5 | -0.75 | 48.98 5452442.4 191975.9 5370350.4 -22.6 | 23518.1 190.6 40467.5 0.721 88.2 1.8 | | 1120.80 | 9370.0 | 26719.0 8.2 33.7 | 0.20 | 15000.0 14999.9 53.0 | 18.19 15.70 -9.19 | 5.94 10232.910232.1 -127.8 | -0.72 | 49.07 5462367.9 191844.8 5379452.1 -22.7 | 23971.3 190.6 41373.9 0.726 88.3 1.7 | | 1121.05 | 9361.7 | 26727.3 0.0 33.7 | 0.20 | 15000.0 14999.9 52.4 | 18.20 15.71 -9.19 | 5.94 10236.710236.0 -126.1 | -0.71 | 49.09 5464796.7 191813.8 5381679.3 -22.7 | 24085.2 190.6 41601.6 0.727 88.3 1.7 | 186

Appendix B2 Sample TRAJ2DF Data Files

Traj2d - Third Stage Coast to Perigee

Case Study: Hypothetical Launch Vehicle

| Time | Vehicle | Acceleration | Velocity | FPA | Distance from Pad | | | Mass | A Ay | Mach V Vx Vy | | DRA SubV Alt Range Elev | | sec | kg | m/s2 m/s2 | No m/s m/s m/s | deg | deg m m m deg | | 1121.05 | 9361.7 | -9.2 -9.2 | 5.94 10236.710235.96 -126.1 | -0.71 | 49.09 5464796.7 191813.8 5381679.3 -22.7 | | 1122.05 | 9361.7 | -9.2 -9.2 | 5.94 10236.910236.15 -119.4 | -0.67 | 49.18 5474734.0 191691.0 5390791.3 -22.7 | | 1123.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.010236.33 -112.7 | -0.63 | 49.27 5484671.7 191575.0 5399903.3 -22.8 | | 1124.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.110236.51 -106.0 | -0.59 | 49.36 5494609.7 191465.6 5409015.2 -22.8 | | 1125.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.110236.67 -99.3 | -0.56 | 49.45 5504548.0 191363.0 5418127.1 -22.9 | | 1126.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.210236.81 -92.6 | -0.52 | 49.54 5514486.7 191267.0 5427238.7 -22.9 | | 1127.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.310236.95 -85.9 | -0.48 | 49.63 5524425.6 191177.8 5436350.3 -23.0 | | 1128.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.410237.08 -79.2 | -0.44 | 49.72 5534364.7 191095.3 5445461.7 -23.0 | | 1129.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.510237.20 -72.5 | -0.41 | 49.81 5544304.1 191019.4 5454573.0 -23.1 | | 1130.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.510237.31 -65.8 | -0.37 | 49.89 5554243.8 190950.3 5463684.1 -23.1 | | 1131.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.610237.41 -59.0 | -0.33 | 49.98 5564183.6 190887.9 5472795.1 -23.2 | | 1132.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.610237.49 -52.3 | -0.29 | 50.07 5574123.6 190832.3 5481905.8 -23.2 | | 1133.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.710237.57 -45.6 | -0.26 | 50.16 5584063.8 190783.3 5491016.4 -23.3 | | 1134.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.710237.63 -38.9 | -0.22 | 50.25 5594004.1 190741.0 5500126.7 -23.3 | | 1135.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.710237.69 -32.2 | -0.18 | 50.34 5603944.5 190705.5 5509236.9 -23.4 | | 1136.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.810237.74 -25.5 | -0.14 | 50.43 5613885.0 190676.6 5518346.8 -23.4 | | 1137.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.810237.77 -18.8 | -0.11 | 50.52 5623825.6 190654.5 5527456.4 -23.5 | | 1138.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.810237.79 -12.1 | -0.07 | 50.61 5633766.2 190639.0 5536565.8 -23.5 | | 1139.05 | 9361.7 | -9.2 -9.2 | 5.94 10237.810237.81 -5.4 | -0.03 | 50.70 5643706.9 190630.3 5545674.9 -23.6 | | 1139.85 | 9361.7 | -9.2 -9.2 | 5.94 10237.810237.81 0.0 | 0.00 | 50.77 5651649.5 190628.2 5552952.9 -23.6 |

187

Appendix B2 Sample TRAJ2DF Data Files

Traj2dF - Flight Summary

Case Study: Hypothetical Launch Vehicle

FLIGHT SUMMARY - Mission Event Listing

FIRST STAGE (Launch Site Relative Frame) Start Pitch Over Altitude 86.5 m Start Pitch Over Velocity 24.6 m/sec End Pitch Over Altitude 3304.4 m End Pitch Over Velocity 193.1 m/sec End Pitch Over Downrange 303.3 m

Max Q Time 66.3 sec Max Q Altitude 11019.5 m, (36153 feet) Max Q Velocity 440.8 m/sec Max Q Mach No 1.5 Max Q Downrange 2939.0 m

S1 Burn Out Time 125.9 sec S1 Burn Out Altitude 57342.0 m, (188129 feet) S1 Burn Out Velocity 1866.8 m/sec S1 Burn Out Mach No 5.8 S1 Burn Out Downrange Distance 42033.6 m

S1 Impact Time 442.6 sec S1 Impact Range 464.3 km

SECOND STAGE (Inertial Frame, No Rentry Drag) S2 Burn Out Time 292.4 sec S2 Burn Out Altitude 271894.0 m, (892040 feet) S2 Burn Out Velocity 4430.1 m/sec S2 Burn Out Downrange Distance 478505.5 m

S2 Impact Time 906.6 sec S2 Impact Range 2872.8 km

FAIRING (Inertial Frame, No Rentry Drag) Fairing Impact Time 523.1 sec Fairing Impact Range 858.2 km

ORBIT ATTAINED (Inertial Frame) Mission Elapsed Time 976.0 sec Mission Altitude 260534.7 m Mission Downrange Distance 4172.4 km

THIRD STAGE (Inertial Frame) S3 Ignition Time 327.8 sec S3 Ignition Altitude 320055.0 m S3 Ignition Velocity 4331.7 m/sec S3 Ignition Downrange Distance 619.3 km S3 Burn Out Time 1121.0 sec S3 Burn Out Altitude 191813.8 m S3 Burn Out Velocity 10236.7 m/sec S3 Burn Out Downrange Distance 5464.8 km S3 Burn Out Perigee Height 190.6 km S3 Burn Out Apogee Radius 41601.6 km S3 Perigee Time 1139.8 sec S3 Perigee Downrange Distance 5651.6 km

188

Appendix B3 Sample TRAJ2DF Data Files

Appendix B3: TRAj2DF Output File – Optimisation Mode

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 1 | 346603.7 92363.5 28870.6 | 11.9 | 7.9 11.3 | 34.5 16.0 | 8.2 0.6 | 30.1 | 474105.8 178.3 42147.0 | 17.7 7.5 | | 2 | 346603.4 93946.5 28054.5 | 11.1 | 7.4 11.6 | 45.3 26.1 | 5.4 0.5 | 103.3 | 474872.4 173.2 42147.6 | 22.8 6.9 | | 3 | 347150.2 91594.0 29365.7 | 13.1 | 7.4 11.6 | 40.4 23.7 | 6.4 1.9 | 66.9 | 474377.9 203.8 42158.4 | 0.0 0.0 | | 4 | 343867.0 89664.5 28000.0 | 10.5 | 7.0 11.1 | 33.5 21.4 | 6.9 2.2 | 67.2 | 467799.5 198.9 42155.6 | 0.0 0.0 | | 5 | 344848.5 90750.8 28606.7 | 10.4 | 7.1 10.1 | 33.3 28.8 | 7.2 2.0 | 76.7 | 470474.0 201.6 42157.2 | 0.0 0.0 | | 6 | 344594.6 89738.8 28619.4 | 10.7 | 7.1 10.5 | 36.7 22.1 | 6.3 1.9 | 63.3 | 469220.8 196.3 42156.0 | 0.0 0.0 | | 7 | 346774.0 90879.6 28623.8 | 10.3 | 7.5 10.8 | 39.6 19.3 | 8.6 3.9 | 70.9 | 472545.4 201.5 42169.2 | 0.0 0.0 | | 8 | 344124.6 89398.2 28000.0 | 10.2 | 7.2 10.2 | 31.3 14.7 | 9.6 1.7 | 35.5 | 467790.8 201.4 42168.9 | 0.0 0.0 | | 9 | 343294.3 93055.4 28034.8 | 12.3 | 7.4 10.8 | 34.9 12.8 | 6.8 3.4 | 30.1 | 470652.5 196.4 42158.6 | 0.0 0.0 | | 10 | 346317.0 91925.1 28168.8 | 10.6 | 7.6 10.4 | 42.5 14.1 | 5.8 1.5 | 49.7 | 472678.8 196.1 42154.9 | 0.0 0.0 | | 11 | 348644.0 92072.6 29331.6 | 10.9 | 7.9 10.5 | 45.0 17.3 | 7.3 1.1 | 49.3 | 476316.2 197.0 42154.5 | 0.0 0.0 | | 12 | 346741.4 91763.2 30493.1 | 11.4 | 7.6 10.9 | 45.9 18.9 | 9.2 3.8 | 45.8 | 475265.7 197.8 42160.0 | 0.0 0.0 | | 13 | 344338.6 92631.3 30507.0 | 11.4 | 7.6 10.4 | 32.6 27.6 | 9.5 2.0 | 33.2 | 473744.9 196.1 42158.2 | 0.0 0.0 | | 14 | 344175.2 92988.0 29055.8 | 12.1 | 7.1 11.8 | 42.9 24.0 | 8.9 2.5 | 85.3 | 472487.0 198.3 42172.3 | 0.0 0.0 | | 15 | 343991.4 90441.4 28849.1 | 10.1 | 7.2 10.8 | 35.8 21.4 | 9.1 0.6 | 53.3 | 469549.9 200.9 42164.6 | 0.0 0.0 | | 16 | 348391.2 91407.4 29035.7 | 10.4 | 7.9 10.9 | 32.2 21.6 | 7.4 4.8 | 42.5 | 475102.3 199.0 42157.7 | 0.0 0.0 | | 17 | 344277.8 93258.9 29604.1 | 13.2 | 7.5 11.4 | 33.8 24.3 | 8.9 0.1 | 43.3 | 473408.8 199.0 42173.4 | 0.0 0.0 | | 18 | 348626.1 91671.3 29083.3 | 13.3 | 7.4 10.4 | 41.0 27.2 | 6.6 2.6 | 92.9 | 475648.7 200.5 42158.8 | 0.0 0.0 | | 19 | 343291.6 92600.0 28382.7 | 12.3 | 7.1 10.5 | 41.0 21.6 | 5.6 0.9 | 72.0 | 470542.3 199.3 42155.7 | 0.0 0.0 | | 20 | 348721.3 93248.3 29776.5 | 13.8 | 7.5 11.8 | 44.8 26.8 | 7.3 4.1 | 87.2 | 478014.1 194.1 42152.8 | 1.9 1.7 | | 21 | 343464.5 90114.2 28057.0 | 10.7 | 7.0 10.0 | 35.5 21.1 | 6.3 0.3 | 61.2 | 467903.7 200.9 42154.6 | 0.0 0.0 | | 22 | 347818.5 90115.0 28167.8 | 12.3 | 7.2 10.6 | 33.8 30.0 | 7.9 1.7 | 112.4 | 472369.3 196.0 42159.3 | 0.0 0.0 | | 23 | 346693.1 89431.3 29886.1 | 11.4 | 7.0 10.9 | 45.8 24.3 | 8.9 3.2 | 78.6 | 472278.5 204.0 42173.7 | 0.0 0.0 | | 24 | 345083.8 89000.0 30080.2 | 10.9 | 7.2 11.3 | 38.5 21.6 | 8.5 2.7 | 43.6 | 470432.0 200.8 42163.4 | 0.0 0.0 | | 25 | 342548.9 92822.2 29409.6 | 10.7 | 7.4 11.0 | 30.9 27.2 | 7.4 1.2 | 36.6 | 471048.7 197.3 42168.6 | 0.0 0.0 | | 26 | 346833.0 90053.0 28441.5 | 10.3 | 7.5 10.0 | 41.8 13.1 | 8.8 0.6 | 50.4 | 471595.5 201.8 42157.1 | 0.0 0.0 | | 27 | 346205.7 90627.3 28575.7 | 10.5 | 7.6 11.6 | 30.7 20.8 | 7.2 1.5 | 37.4 | 471676.8 196.1 42164.2 | 0.0 0.0 | | 28 | 345918.4 92219.0 28088.4 | 12.0 | 7.3 10.9 | 37.9 23.5 | 5.6 1.8 | 82.6 | 472493.8 197.2 42155.8 | 0.0 0.0 | | 29 | 348647.7 90799.8 29521.5 | 10.5 | 7.5 10.4 | 44.2 23.9 | 8.9 4.6 | 81.9 | 475237.0 196.3 42156.6 | 0.0 0.0 |

189

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 30 | 344604.5 93159.9 28390.7 | 10.6 | 7.7 11.0 | 32.1 20.2 | 7.1 0.0 | 30.0 | 472423.0 197.1 42163.0 | 0.0 0.0 | | 31 | 344023.8 93141.8 29011.6 | 10.7 | 7.6 10.5 | 30.7 26.5 | 6.9 0.8 | 38.2 | 472445.2 200.0 42160.5 | 0.0 0.0 | | 32 | 348105.7 92680.0 28401.7 | 10.0 | 7.8 10.5 | 36.2 24.2 | 9.0 3.5 | 80.7 | 475455.5 196.8 42162.9 | 0.0 0.0 | | 33 | 348534.5 93740.1 28992.1 | 13.0 | 7.7 11.7 | 35.9 28.3 | 9.8 2.7 | 93.1 | 477534.7 197.8 42154.9 | 0.0 0.0 | | 34 | 348069.9 92474.3 29318.7 | 11.0 | 7.7 12.0 | 45.6 17.4 | 7.8 3.9 | 59.5 | 476130.9 200.6 42154.6 | 0.0 0.0 | | 35 | 344716.5 91263.9 28371.9 | 12.0 | 7.1 11.1 | 36.1 26.8 | 6.3 2.1 | 88.6 | 470620.2 198.7 42158.2 | 0.0 0.0 | | 36 | 346045.2 93948.6 28668.6 | 12.1 | 7.8 11.9 | 39.7 13.8 | 6.0 4.6 | 34.0 | 474930.4 200.9 42168.9 | 0.0 0.0 | | 37 | 344880.2 92218.1 28221.1 | 12.5 | 7.3 10.0 | 33.1 24.3 | 6.4 1.4 | 68.8 | 471587.4 202.8 42155.3 | 0.0 0.0 | | 38 | 346090.6 89690.0 29607.5 | 11.2 | 7.4 11.5 | 38.1 20.9 | 7.4 0.6 | 42.3 | 471656.0 196.5 42157.1 | 0.0 0.0 | | 39 | 348827.0 91763.7 29004.7 | 10.5 | 8.1 11.1 | 38.1 14.9 | 7.9 3.5 | 33.5 | 475863.4 196.8 42170.5 | 0.0 0.0 | | 40 | 346672.4 90513.4 28601.6 | 10.5 | 7.6 11.0 | 37.1 16.9 | 5.0 2.3 | 34.7 | 472055.4 196.8 42155.0 | 0.0 0.0 | | 41 | 347218.7 91610.9 28749.7 | 13.5 | 7.2 11.5 | 37.2 27.7 | 7.9 4.8 | 104.9 | 473847.3 203.1 42158.4 | 0.0 0.0 | | 42 | 346884.4 92056.6 30081.3 | 12.3 | 7.5 10.3 | 42.7 24.5 | 7.2 1.5 | 53.7 | 475290.3 202.8 42154.6 | 0.0 0.0 | | 43 | 347526.6 91698.5 28655.1 | 11.9 | 7.4 10.2 | 43.7 20.7 | 9.0 0.1 | 85.2 | 474148.2 200.3 42159.0 | 0.0 0.0 | | 44 | 345748.0 89743.8 29009.5 | 10.8 | 7.2 10.7 | 42.6 18.3 | 6.1 3.1 | 59.7 | 470769.3 196.6 42160.9 | 0.0 0.0 | | 45 | 344078.8 90248.0 29561.1 | 13.8 | 7.2 11.0 | 32.6 21.4 | 7.1 0.7 | 33.7 | 470155.9 198.0 42156.6 | 0.0 0.0 | | 46 | 346615.0 92623.5 28119.3 | 10.9 | 7.4 10.0 | 37.6 27.5 | 5.4 0.0 | 75.6 | 473625.9 217.9 42184.2 | 13.9 9.7 | | 47 | 345687.2 93053.5 28463.8 | 12.1 | 7.7 12.0 | 33.6 17.9 | 5.8 3.3 | 34.5 | 473472.5 197.9 42167.9 | 0.0 0.0 | | 48 | 348212.2 89456.9 30928.7 | 11.5 | 7.7 10.7 | 33.1 28.0 | 9.5 0.5 | 32.3 | 474865.8 197.1 42157.3 | 0.0 0.0 | | 49 | 348110.1 92488.0 30131.3 | 12.1 | 7.9 10.6 | 39.1 21.2 | 6.3 4.6 | 33.6 | 476997.4 203.6 42155.9 | 0.0 0.0 | | 50 | 348218.4 92586.7 29161.5 | 14.0 | 7.6 11.5 | 38.7 25.3 | 8.8 1.8 | 86.3 | 476234.5 198.4 42167.8 | 0.0 0.0 | | 51 | 346452.8 89400.6 30106.8 | 12.8 | 7.4 11.8 | 40.4 18.0 | 9.6 0.8 | 42.2 | 472228.1 201.7 42165.4 | 0.0 0.0 | | 52 | 343677.8 92412.4 28365.3 | 13.8 | 7.2 10.4 | 34.7 19.8 | 5.8 0.5 | 49.2 | 470723.6 202.4 42154.5 | 0.0 0.0 | | 53 | 347848.0 89389.6 29845.3 | 13.4 | 7.0 10.4 | 47.2 22.8 | 8.5 1.5 | 85.9 | 473350.9 196.2 42166.3 | 0.0 0.0 | | 54 | 345470.5 89516.2 28370.5 | 10.2 | 7.1 10.5 | 35.7 26.0 | 5.2 2.6 | 76.8 | 469625.2 197.3 42161.4 | 0.0 0.0 | | 55 | 344033.7 89994.3 28758.0 | 10.3 | 7.2 10.0 | 39.9 15.1 | 5.8 0.3 | 31.8 | 469054.0 187.9 42150.7 | 8.1 3.8 | | 56 | 342015.0 93917.6 28604.1 | 10.2 | 7.1 11.6 | 43.7 21.4 | 6.5 2.1 | 60.8 | 470804.7 209.3 42173.6 | 5.3 0.0 | | 57 | 346153.9 90557.4 29548.5 | 11.2 | 7.3 11.1 | 45.4 17.9 | 8.1 2.1 | 56.8 | 472527.9 201.7 42155.8 | 0.0 0.0 | | 58 | 346348.1 93895.7 28269.8 | 13.4 | 7.6 11.7 | 35.0 23.7 | 5.4 2.7 | 71.3 | 474781.7 198.1 42164.6 | 0.0 0.0 | | 59 | 343241.6 90061.7 29333.8 | 10.1 | 7.0 10.5 | 33.2 28.5 | 8.6 1.0 | 59.5 | 468905.1 198.9 42155.1 | 0.0 0.0 | | 60 | 347078.5 90367.5 28571.9 | 11.1 | 7.7 12.0 | 30.9 17.2 | 7.7 2.5 | 30.0 | 472285.8 196.0 42154.6 | 0.0 0.0 | | 61 | 344680.6 89319.6 28603.3 | 11.7 | 7.2 10.0 | 30.1 17.6 | 7.6 3.5 | 35.3 | 468871.5 198.3 42163.5 | 0.0 0.0 | | 62 | 344079.4 93276.8 28000.0 | 11.1 | 7.5 10.2 | 37.9 16.9 | 7.1 1.5 | 52.8 | 471624.2 198.3 42169.6 | 0.0 0.0 | | 63 | 346317.9 89542.9 28304.7 | 11.7 | 7.2 11.3 | 38.9 14.0 | 9.9 4.3 | 70.7 | 470433.5 202.9 42157.1 | 0.0 0.0 |

190

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 64 | 348366.9 93329.1 30866.2 | 13.5 | 7.8 11.3 | 43.7 23.5 | 7.6 5.0 | 47.7 | 478830.2 197.4 42156.9 | 0.0 0.0 | | 65 | 348889.6 91362.0 29929.7 | 11.8 | 7.5 10.5 | 49.1 21.2 | 7.5 2.4 | 67.2 | 476449.3 196.3 42154.6 | 0.0 0.0 | | 66 | 347806.2 89679.9 29658.4 | 11.3 | 7.4 11.1 | 43.6 21.3 | 8.9 1.6 | 68.2 | 473412.5 200.4 42167.9 | 0.0 0.0 | | 67 | 344446.7 92766.4 28034.9 | 11.4 | 7.5 10.0 | 31.7 21.7 | 6.3 2.2 | 52.4 | 471516.0 201.6 42157.8 | 0.0 0.0 | | 68 | 347287.0 89736.1 28556.3 | 10.9 | 7.4 11.7 | 38.0 22.0 | 7.9 2.0 | 78.4 | 471847.4 196.4 42160.3 | 0.0 0.0 | | 69 | 344400.4 89995.0 28314.4 | 10.2 | 7.3 10.9 | 33.1 19.6 | 5.3 0.2 | 32.8 | 468977.7 199.2 42162.2 | 0.0 0.0 | | 70 | 347392.8 92557.1 29203.4 | 12.3 | 7.7 10.5 | 31.5 26.3 | 9.7 2.3 | 64.1 | 475421.3 201.1 42170.5 | 0.0 0.0 | | 71 | 346887.0 93286.0 29825.4 | 12.8 | 7.5 11.0 | 43.3 25.0 | 9.9 3.3 | 80.5 | 476266.5 196.2 42157.3 | 0.0 0.0 | | 72 | 347505.6 92264.0 29926.6 | 12.7 | 7.3 11.2 | 48.6 26.5 | 5.7 0.9 | 66.1 | 475964.2 200.8 42165.2 | 0.0 0.0 | | 73 | 345538.6 89449.4 29102.2 | 10.2 | 7.2 10.5 | 34.4 26.2 | 7.6 1.0 | 60.6 | 470358.3 198.3 42157.8 | 0.0 0.0 | | 74 | 344474.4 92595.6 29661.4 | 12.3 | 7.1 12.0 | 47.1 23.0 | 8.3 1.0 | 74.7 | 472999.4 186.2 42144.7 | 9.8 9.8 | | 75 | 347618.0 92849.8 29364.9 | 10.5 | 7.6 10.1 | 46.0 21.4 | 8.8 4.0 | 73.6 | 476100.7 198.1 42155.4 | 0.0 0.0 | | 76 | 345165.1 91146.5 28208.0 | 11.7 | 7.3 11.1 | 34.7 20.3 | 5.7 3.9 | 63.0 | 470787.6 196.2 42155.1 | 0.0 0.0 | | 77 | 347863.9 89979.4 28451.7 | 11.5 | 7.6 10.1 | 38.1 17.2 | 7.3 1.4 | 61.5 | 472562.9 190.7 42149.4 | 5.3 5.1 | | 78 | 348527.6 89176.0 29195.8 | 10.8 | 7.1 10.0 | 48.7 23.8 | 7.0 2.9 | 89.5 | 473167.5 200.4 42160.1 | 0.0 0.0 | | 79 | 348340.4 93403.4 28266.0 | 12.3 | 7.3 10.4 | 45.7 26.9 | 9.8 4.0 | 129.4 | 476277.8 196.3 42157.6 | 0.0 0.0 | | 80 | 346010.0 89000.0 28427.0 | 11.0 | 7.3 10.6 | 31.1 21.3 | 5.5 4.5 | 52.1 | 469705.1 200.8 42163.7 | 0.0 0.0 | | 81 | 345854.9 92952.1 28617.2 | 11.4 | 7.8 10.0 | 31.7 17.9 | 9.9 3.1 | 40.4 | 473692.1 196.2 42162.2 | 0.0 0.0 | | 82 | 347842.0 89564.1 28489.7 | 13.7 | 7.4 10.1 | 35.1 15.8 | 9.0 1.3 | 59.1 | 472163.8 196.6 42154.7 | 0.0 0.0 | | 83 | 346680.4 89700.2 29215.5 | 11.6 | 7.3 10.5 | 32.1 29.1 | 7.2 0.7 | 66.8 | 471864.1 196.4 42154.8 | 0.0 0.0 | | 84 | 345615.3 89452.7 30697.1 | 12.3 | 7.1 11.0 | 42.6 26.1 | 9.2 0.4 | 56.5 | 472033.2 196.8 42154.8 | 0.0 0.0 | | 85 | 345916.4 93046.4 28615.3 | 12.1 | 7.6 10.3 | 34.1 24.3 | 7.2 0.8 | 61.2 | 473846.1 198.0 42155.7 | 0.0 0.0 | | 86 | 343976.6 90461.6 28116.1 | 11.4 | 7.2 10.3 | 30.4 20.9 | 6.0 0.2 | 42.1 | 468822.3 198.2 42158.1 | 0.0 0.0 | | 87 | 342108.1 92474.2 28030.6 | 13.0 | 7.1 10.3 | 30.7 17.4 | 9.0 2.5 | 46.0 | 468881.0 203.1 42168.9 | 0.0 0.0 | | 88 | 342982.3 89032.1 29516.8 | 10.5 | 7.0 10.4 | 36.4 21.9 | 6.8 0.3 | 35.1 | 467799.2 196.1 42162.6 | 0.0 0.0 | | 89 | 343342.7 90819.8 28747.9 | 11.8 | 7.0 10.1 | 43.8 14.8 | 7.2 0.7 | 53.9 | 469178.4 199.1 42168.4 | 0.0 0.0 | | 90 | 348301.8 93096.3 28951.6 | 10.6 | 8.1 10.4 | 43.0 11.5 | 9.7 0.9 | 32.6 | 476617.7 199.7 42157.3 | 0.0 0.0 | | 91 | 343170.7 92328.8 28151.0 | 12.5 | 7.0 11.1 | 43.2 16.1 | 6.3 3.1 | 73.8 | 469918.4 201.0 42156.2 | 0.0 0.0 | | 92 | 346130.6 92669.9 28991.6 | 11.4 | 7.8 11.9 | 39.7 15.2 | 6.8 1.9 | 30.4 | 474060.1 199.6 42163.1 | 0.0 0.0 | | 93 | 345520.8 92377.0 28107.9 | 10.2 | 7.6 10.4 | 36.5 18.6 | 5.5 1.4 | 43.7 | 472273.7 199.1 42154.6 | 0.0 0.0 | | 94 | 343958.9 90421.3 28200.2 | 10.0 | 7.2 11.0 | 41.6 11.1 | 5.9 1.7 | 33.1 | 468848.4 198.0 42164.1 | 0.0 0.0 | | 95 | 346249.0 90795.4 28000.0 | 12.4 | 7.4 10.8 | 30.9 22.2 | 7.3 2.8 | 73.1 | 471312.4 197.6 42161.5 | 0.0 0.0 | | 96 | 344567.3 92241.0 28012.3 | 11.8 | 7.4 10.1 | 30.1 22.6 | 7.5 2.5 | 61.9 | 471088.6 200.4 42154.5 | 0.0 0.0 | | 97 | 342948.3 90151.6 28091.1 | 10.6 | 7.1 10.4 | 36.0 14.4 | 6.2 1.5 | 34.5 | 467459.0 196.0 42166.3 | 0.0 0.0 |

191

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 98 | 344100.2 91667.0 28346.3 | 12.0 | 7.4 10.5 | 33.4 17.7 | 8.3 1.0 | 44.7 | 470381.4 196.5 42161.9 | 0.0 0.0 | | 99 | 342703.8 92623.3 28705.1 | 11.7 | 7.4 10.0 | 31.9 20.1 | 7.0 1.9 | 31.5 | 470300.2 202.4 42157.1 | 0.0 0.0 | | 100 | 343862.0 91629.1 29075.6 | 11.1 | 7.1 10.9 | 37.6 28.2 | 5.5 2.0 | 67.7 | 470834.8 198.5 42163.9 | 0.0 0.0 | | 101 | 343125.6 90887.3 28000.0 | 12.2 | 7.1 11.3 | 30.0 20.9 | 5.1 1.4 | 44.6 | 468281.0 197.0 42158.7 | 0.0 0.0 | | 102 | 345950.7 90618.9 28333.0 | 10.5 | 7.5 10.3 | 36.4 17.8 | 5.8 1.1 | 42.5 | 471170.7 196.1 42159.5 | 0.0 0.0 | | 103 | 347907.8 91628.7 28151.2 | 13.7 | 7.5 10.8 | 36.8 19.9 | 5.9 0.5 | 69.5 | 473955.8 198.4 42155.1 | 0.0 0.0 | | 104 | 344295.5 90532.4 28283.6 | 10.0 | 7.3 10.1 | 37.9 18.3 | 5.5 0.1 | 43.4 | 469379.5 196.3 42157.0 | 0.0 0.0 | | 105 | 344524.2 90035.0 28489.8 | 11.7 | 7.0 11.4 | 35.6 25.1 | 6.6 1.2 | 76.9 | 469316.9 203.1 42162.0 | 0.0 0.0 | | 106 | 348544.1 92970.0 30538.1 | 13.6 | 7.5 11.4 | 45.5 28.4 | 7.3 4.5 | 70.4 | 478320.2 200.4 42159.6 | 0.0 0.0 | | 107 | 346897.5 92211.6 30667.9 | 11.1 | 7.8 10.4 | 38.3 26.0 | 8.4 0.8 | 31.3 | 476045.0 196.6 42172.3 | 0.0 0.0 | | 108 | 343945.1 89801.8 28370.7 | 10.5 | 7.1 11.7 | 39.8 14.8 | 5.4 2.8 | 45.3 | 468385.6 197.0 42170.7 | 0.0 0.0 | | 109 | 345803.2 91284.8 28090.3 | 12.8 | 7.3 11.8 | 41.0 15.1 | 5.2 0.2 | 54.1 | 471446.3 203.7 42155.0 | 0.0 0.0 | | 110 | 345265.7 92663.8 28704.9 | 10.8 | 7.1 11.6 | 43.8 29.4 | 7.2 0.1 | 92.3 | 472902.3 194.2 42152.8 | 1.8 1.7 | | 111 | 343733.2 89895.0 28010.1 | 10.3 | 7.2 11.2 | 33.9 18.3 | 6.7 0.3 | 45.5 | 467906.3 198.2 42155.3 | 0.0 0.0 | | 112 | 342683.0 93746.7 28598.5 | 10.2 | 7.6 11.6 | 34.3 18.4 | 9.1 0.3 | 30.0 | 471296.2 196.1 42162.8 | 0.0 0.0 | | 113 | 344977.1 91769.1 28384.9 | 11.9 | 7.5 11.2 | 35.1 17.2 | 7.4 1.4 | 43.9 | 471399.1 200.6 42156.5 | 0.0 0.0 | | 114 | 348863.8 92547.3 28107.9 | 11.2 | 8.2 11.1 | 31.5 15.3 | 9.5 0.4 | 30.0 | 475787.0 196.5 42155.2 | 0.0 0.0 | | 115 | 346613.0 92154.2 28979.2 | 10.7 | 7.7 10.7 | 44.3 11.4 | 8.9 2.3 | 39.5 | 474014.4 197.6 42154.6 | 0.0 0.0 | | 116 | 347971.2 89446.0 28055.2 | 12.9 | 7.3 11.2 | 44.1 12.1 | 5.1 0.3 | 61.3 | 471740.4 200.3 42162.5 | 0.0 0.0 | | 117 | 343747.9 90580.0 28037.1 | 11.7 | 7.2 11.1 | 39.9 10.3 | 6.3 1.7 | 36.7 | 468632.9 196.1 42162.3 | 0.0 0.0 | | 118 | 344395.6 91346.4 30801.9 | 11.5 | 7.1 11.2 | 45.6 27.1 | 5.9 3.3 | 42.0 | 472811.9 203.6 42168.4 | 0.0 0.0 | | 119 | 348632.9 91168.1 28758.5 | 11.1 | 8.0 11.9 | 34.8 15.2 | 9.2 0.2 | 30.1 | 474827.5 194.2 42161.4 | 1.8 0.0 | | 120 | 348952.5 89601.9 29827.4 | 10.1 | 7.5 12.0 | 49.4 17.4 | 9.1 0.1 | 54.8 | 474649.8 198.1 42160.0 | 0.0 0.0 | | 121 | 344665.2 91687.8 28823.8 | 12.9 | 7.4 11.1 | 35.0 16.0 | 8.1 4.8 | 43.5 | 471444.8 200.9 42157.2 | 0.0 0.0 | | 122 | 347503.9 90677.4 28653.7 | 10.4 | 7.2 10.6 | 44.8 23.4 | 7.8 5.0 | 98.5 | 473103.1 198.0 42162.9 | 0.0 0.0 | | 123 | 344114.6 92285.9 28673.6 | 10.3 | 7.4 10.5 | 42.5 16.5 | 5.3 0.1 | 35.2 | 471342.0 198.1 42167.7 | 0.0 0.0 | | 124 | 342576.3 91610.4 28676.7 | 11.9 | 7.2 10.4 | 38.7 15.7 | 5.8 0.8 | 33.0 | 469131.4 196.2 42155.1 | 0.0 0.0 | | 125 | 342641.4 91782.0 28771.1 | 10.8 | 7.0 10.1 | 32.7 29.9 | 5.2 0.1 | 57.7 | 469462.4 199.0 42157.3 | 0.0 0.0 | | 126 | 344067.2 89028.9 29382.6 | 11.2 | 7.2 10.2 | 38.6 14.2 | 9.6 1.5 | 34.7 | 468746.7 197.4 42166.5 | 0.0 0.0 | | 127 | 342066.9 92099.9 28055.8 | 10.1 | 7.0 12.0 | 37.7 23.0 | 5.3 1.0 | 65.3 | 468490.6 196.2 42154.7 | 0.0 0.0 | | 128 | 346934.3 90625.5 28368.8 | 12.0 | 7.6 11.0 | 35.6 14.0 | 7.7 3.4 | 44.3 | 472196.6 196.0 42155.1 | 0.0 0.0 | | 129 | 344005.2 91528.4 28388.7 | 11.7 | 7.2 11.9 | 43.9 12.9 | 7.5 2.5 | 65.8 | 470190.3 179.1 42154.9 | 16.9 0.0 | | 130 | 343176.8 90799.7 28825.1 | 10.6 | 7.0 10.8 | 41.8 19.0 | 6.4 1.2 | 53.3 | 469069.5 200.5 42162.3 | 0.0 0.0 | | 131 | 345917.8 89927.2 28000.0 | 10.2 | 7.4 10.0 | 38.5 10.7 | 9.2 0.5 | 39.5 | 470112.9 196.1 42155.0 | 0.0 0.0 |

192

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 132 | 343490.1 90552.7 29510.3 | 11.3 | 7.1 11.1 | 45.5 11.8 | 7.6 1.4 | 30.2 | 469821.1 197.0 42155.5 | 0.0 0.0 | | 133 | 342899.4 91982.8 28107.7 | 12.3 | 7.1 10.4 | 39.9 11.9 | 6.5 4.1 | 47.9 | 469257.8 197.8 42157.4 | 0.0 0.0 | | 134 | 348402.1 89104.3 30812.7 | 12.6 | 7.4 11.5 | 44.1 21.7 | 9.9 0.0 | 48.8 | 474587.1 200.2 42159.4 | 0.0 0.0 | | 135 | 348295.5 92362.7 28008.5 | 13.6 | 7.6 10.2 | 34.6 19.0 | 9.1 4.1 | 84.4 | 474934.7 196.4 42165.8 | 0.0 0.0 | | 136 | 346147.8 93793.5 29012.0 | 10.9 | 7.4 10.3 | 49.3 19.4 | 6.7 3.0 | 68.0 | 475221.3 204.3 42174.5 | 0.3 0.0 | | 137 | 345583.5 89249.2 28052.9 | 10.0 | 7.2 10.6 | 40.7 15.8 | 7.3 0.0 | 61.8 | 469153.6 196.5 42155.4 | 0.0 0.0 | | 138 | 347124.2 91451.3 28880.4 | 10.2 | 7.9 11.1 | 32.4 18.7 | 9.0 1.5 | 30.9 | 473723.8 196.6 42156.6 | 0.0 0.0 | | 139 | 348306.5 89206.8 28324.4 | 11.2 | 7.5 11.5 | 42.4 11.3 | 5.1 2.5 | 45.9 | 472105.7 196.5 42168.3 | 0.0 0.0 | | 140 | 348307.5 92852.0 30272.8 | 11.8 | 7.9 11.5 | 41.7 21.2 | 9.4 3.9 | 49.0 | 477700.3 199.6 42158.1 | 0.0 0.0 | | 141 | 343874.6 90790.7 30902.7 | 11.6 | 7.2 11.3 | 34.3 29.4 | 7.1 2.3 | 30.0 | 471836.0 195.1 42156.9 | 0.9 0.0 | | 142 | 348970.5 92524.5 28852.8 | 12.1 | 8.0 10.3 | 40.5 15.5 | 7.3 0.8 | 42.5 | 476615.8 199.3 42173.0 | 0.0 0.0 | | 143 | 348785.2 91021.8 30976.2 | 13.2 | 7.8 10.3 | 36.8 27.3 | 7.0 2.3 | 35.2 | 477051.3 198.3 42154.7 | 0.0 0.0 | | 144 | 347410.4 89797.8 30244.6 | 10.6 | 7.5 11.0 | 36.7 24.5 | 9.7 4.5 | 53.0 | 473720.7 196.9 42158.1 | 0.0 0.0 | | 145 | 348831.4 90607.1 28542.6 | 13.0 | 7.3 10.2 | 40.7 23.5 | 9.4 4.3 | 121.0 | 474249.0 172.7 42138.0 | 23.3 16.5 | | 146 | 345572.7 89670.5 28697.1 | 13.3 | 7.2 11.5 | 31.9 19.6 | 9.2 0.1 | 53.9 | 470208.4 197.2 42172.8 | 0.0 0.0 | | 147 | 346545.7 91287.0 30247.0 | 13.3 | 7.4 10.1 | 34.4 27.9 | 6.7 3.2 | 50.0 | 474347.6 197.1 42157.5 | 0.0 0.0 | | 148 | 342187.8 93770.0 29386.9 | 13.4 | 7.2 10.7 | 41.1 18.0 | 8.6 1.1 | 46.7 | 471612.7 197.8 42158.9 | 0.0 0.0 | | 149 | 344190.5 90866.7 28688.1 | 11.3 | 7.1 10.8 | 41.0 19.0 | 7.2 3.2 | 68.7 | 470013.2 199.3 42170.0 | 0.0 0.0 | | 150 | 346175.0 89526.5 28196.2 | 11.8 | 7.4 10.3 | 32.6 16.1 | 8.3 3.4 | 52.1 | 470165.7 196.9 42157.5 | 0.0 0.0 | | 151 | 346194.7 90551.7 28190.3 | 11.5 | 7.4 11.0 | 37.9 16.3 | 5.5 2.9 | 53.6 | 471204.8 199.7 42165.9 | 0.0 0.0 | | 152 | 344914.0 93276.4 29039.0 | 13.0 | 7.4 11.5 | 42.4 20.1 | 6.8 0.3 | 58.4 | 473497.4 200.4 42158.9 | 0.0 0.0 | | 153 | 345427.6 91609.4 29077.9 | 10.5 | 7.5 11.0 | 32.7 27.2 | 6.4 1.3 | 49.7 | 472382.9 198.0 42166.0 | 0.0 0.0 | | 154 | 347676.1 93910.1 30539.7 | 12.2 | 7.9 11.3 | 48.2 17.3 | 6.1 4.2 | 30.0 | 478393.9 199.2 42159.7 | 0.0 0.0 | | 155 | 343001.3 89641.3 28763.0 | 12.5 | 7.1 11.3 | 33.4 14.5 | 7.5 0.3 | 30.0 | 467673.6 167.1 42137.8 | 28.9 16.7 | | 156 | 345810.4 90932.1 30327.4 | 12.5 | 7.4 10.4 | 40.5 22.0 | 9.9 1.3 | 48.3 | 473337.8 203.5 42156.5 | 0.0 0.0 | | 157 | 348719.8 93469.1 29046.5 | 11.3 | 7.8 11.2 | 42.1 24.5 | 6.2 2.5 | 70.9 | 477503.3 198.4 42157.2 | 0.0 0.0 | | 158 | 346230.3 89066.0 28429.7 | 12.5 | 7.2 11.6 | 39.8 13.7 | 5.3 0.0 | 41.8 | 469994.0 200.5 42172.2 | 0.0 0.0 | | 159 | 346126.6 93785.6 28985.2 | 13.6 | 7.8 10.2 | 31.7 19.8 | 7.5 3.9 | 37.5 | 475165.4 196.4 42169.8 | 0.0 0.0 | | 160 | 344906.7 92678.0 29018.2 | 11.5 | 7.2 11.1 | 46.7 20.6 | 7.1 0.9 | 70.8 | 472870.9 196.8 42157.5 | 0.0 0.0 | | 161 | 345532.5 91770.5 29178.8 | 10.8 | 7.7 10.5 | 31.7 22.1 | 8.3 0.7 | 32.3 | 472749.8 196.2 42159.8 | 0.0 0.0 | | 162 | 345329.3 90132.1 28966.5 | 11.8 | 7.3 10.1 | 43.5 10.6 | 7.4 0.5 | 30.8 | 470695.9 202.8 42155.5 | 0.0 0.0 | | 163 | 346833.0 89490.9 29498.1 | 10.9 | 7.2 10.8 | 47.9 17.1 | 6.8 0.8 | 55.3 | 472089.9 196.1 42158.6 | 0.0 0.0 | | 164 | 347125.9 89193.9 29243.1 | 10.3 | 7.6 11.8 | 42.8 10.8 | 9.4 0.2 | 30.0 | 471830.9 193.4 42170.0 | 2.6 0.0 | | 165 | 346083.8 90174.2 30286.1 | 11.6 | 7.3 11.9 | 48.4 14.0 | 9.9 0.2 | 40.3 | 472812.1 193.1 42173.7 | 2.9 0.0 |

193

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 166 | 343895.8 90686.4 29476.5 | 10.4 | 7.4 10.0 | 35.9 18.7 | 8.6 2.6 | 30.7 | 470326.7 200.6 42171.2 | 0.0 0.0 | | 167 | 344014.2 90770.0 28171.8 | 11.4 | 7.1 10.3 | 42.9 11.8 | 7.7 0.4 | 51.3 | 469224.0 201.5 42155.4 | 0.0 0.0 | | 168 | 344356.8 90519.4 29296.3 | 10.4 | 7.3 10.3 | 34.7 19.2 | 9.4 4.8 | 44.1 | 470440.5 196.5 42160.7 | 0.0 0.0 | | 169 | 342656.6 91230.4 29672.9 | 12.6 | 7.0 10.2 | 33.0 26.9 | 6.0 4.1 | 51.8 | 469828.0 196.2 42173.4 | 0.0 0.0 | | 170 | 342955.6 93481.2 28791.7 | 10.6 | 7.5 10.0 | 30.2 22.3 | 9.6 3.8 | 43.2 | 471496.5 198.4 42159.9 | 0.0 0.0 | | 171 | 344693.1 93907.0 30603.6 | 13.8 | 7.2 10.1 | 44.7 27.1 | 9.1 2.5 | 63.2 | 475471.6 199.1 42155.4 | 0.0 0.0 | | 172 | 343423.4 92171.8 30821.3 | 11.0 | 7.0 10.7 | 45.0 27.8 | 7.3 4.9 | 45.8 | 472684.5 204.0 42160.0 | 0.0 0.0 | | 173 | 346062.6 92449.5 29078.9 | 12.2 | 7.6 10.1 | 41.1 18.0 | 7.1 0.5 | 47.0 | 473859.0 198.4 42160.0 | 0.0 0.0 | | 174 | 343349.8 93490.1 28359.9 | 13.3 | 7.3 11.0 | 42.7 11.7 | 6.5 0.7 | 39.7 | 471467.8 198.5 42169.0 | 0.0 0.0 | | 175 | 346678.1 89478.2 28906.1 | 11.6 | 7.2 10.6 | 40.9 19.8 | 7.1 3.8 | 75.7 | 471330.4 196.5 42166.0 | 0.0 0.0 | | 176 | 345740.7 91778.0 29837.2 | 13.9 | 7.3 10.9 | 35.2 27.2 | 9.3 2.1 | 68.1 | 473623.9 201.7 42154.6 | 0.0 0.0 | | 177 | 348463.7 89269.1 28372.4 | 13.1 | 7.4 10.1 | 34.7 21.3 | 5.2 1.8 | 69.1 | 472373.2 198.5 42158.9 | 0.0 0.0 | | 178 | 345718.6 89964.1 28227.0 | 13.5 | 7.3 10.5 | 36.6 13.4 | 6.7 0.9 | 44.7 | 470177.6 196.0 42154.7 | 0.0 0.0 | | 179 | 348215.1 90476.9 28183.3 | 11.5 | 7.2 10.7 | 46.9 19.4 | 8.4 1.3 | 105.2 | 473143.2 197.0 42158.2 | 0.0 0.0 | | 180 | 347645.9 89974.2 28552.6 | 10.0 | 7.4 10.1 | 43.9 15.5 | 7.5 2.7 | 63.6 | 472440.8 202.4 42162.7 | 0.0 0.0 | | 181 | 348942.3 91704.2 28293.3 | 13.3 | 7.7 11.0 | 32.1 25.2 | 5.8 1.2 | 73.1 | 475207.8 196.7 42157.1 | 0.0 0.0 | | 182 | 346124.4 92485.1 28206.5 | 12.5 | 7.3 11.8 | 48.2 12.5 | 5.9 2.2 | 74.2 | 473084.0 188.1 42146.6 | 7.9 7.9 | | 183 | 342532.3 89470.8 28053.3 | 10.1 | 7.0 11.2 | 38.9 11.6 | 6.9 0.1 | 31.0 | 466324.5 196.5 42157.8 | 0.0 0.0 | | 184 | 345280.9 91545.3 28505.3 | 11.6 | 7.3 11.4 | 40.5 18.3 | 7.4 3.2 | 70.0 | 471599.5 200.0 42155.5 | 0.0 0.0 | | 185 | 344507.4 93957.8 28301.9 | 13.7 | 7.3 11.5 | 31.8 27.1 | 6.7 3.5 | 83.8 | 473035.1 196.5 42155.5 | 0.0 0.0 | | 186 | 342417.1 90776.0 30319.0 | 11.1 | 7.0 10.3 | 32.9 28.7 | 7.5 3.7 | 40.2 | 469780.2 197.6 42160.4 | 0.0 0.0 | | 187 | 344241.3 89741.1 28292.9 | 12.9 | 7.0 10.0 | 35.1 16.6 | 6.4 4.2 | 56.2 | 468543.2 204.5 42169.9 | 0.5 0.0 | | 188 | 345996.6 91486.2 28609.3 | 10.5 | 7.5 11.4 | 31.3 27.9 | 5.2 1.9 | 58.8 | 472360.0 196.2 42160.7 | 0.0 0.0 | | 189 | 342584.6 93892.4 28000.0 | 12.5 | 7.3 10.0 | 30.4 23.1 | 8.9 3.0 | 70.4 | 470744.9 196.3 42155.8 | 0.0 0.0 | | 190 | 347168.9 89641.1 30427.9 | 13.7 | 7.2 10.2 | 40.7 26.1 | 6.3 2.3 | 58.8 | 473505.8 196.1 42156.4 | 0.0 0.0 | | 191 | 345932.6 89313.2 28801.8 | 12.4 | 7.0 10.7 | 42.1 20.1 | 6.3 1.4 | 74.1 | 470315.6 202.6 42164.8 | 0.0 0.0 | | 192 | 348503.6 89501.6 29016.5 | 11.6 | 7.6 11.6 | 34.6 23.6 | 5.5 1.5 | 52.3 | 473289.7 204.0 42155.8 | 0.0 0.0 | | 193 | 344252.6 90378.3 28201.1 | 12.8 | 7.2 10.1 | 30.5 20.1 | 6.9 0.3 | 47.2 | 469099.9 200.4 42155.7 | 0.0 0.0 | | 194 | 345518.0 89590.0 28001.6 | 10.3 | 7.3 10.2 | 40.6 10.1 | 5.8 3.4 | 39.8 | 469377.6 196.3 42156.3 | 0.0 0.0 | | 195 | 344798.6 92679.5 29613.5 | 11.2 | 7.3 10.6 | 39.0 29.0 | 5.0 2.0 | 48.2 | 473359.5 218.5 42182.4 | 14.5 7.9 | | 196 | 345028.0 89512.7 28783.6 | 10.7 | 7.0 10.4 | 43.0 19.6 | 5.2 1.1 | 59.2 | 469592.3 203.2 42166.5 | 0.0 0.0 | | 197 | 346710.6 91450.7 28534.1 | 10.4 | 7.6 10.3 | 32.4 23.2 | 8.9 2.2 | 62.8 | 472963.4 196.1 42161.1 | 0.0 0.0 | | 198 | 345189.1 90797.3 29178.4 | 11.4 | 7.5 10.1 | 32.3 21.1 | 8.6 2.6 | 42.1 | 471432.8 200.2 42167.9 | 0.0 0.0 | | 199 | 345744.0 93871.0 28357.4 | 10.9 | 7.4 10.5 | 49.3 16.7 | 6.7 2.6 | 80.6 | 474240.5 187.8 42146.7 | 8.2 7.8 |

194

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 200 | 344929.4 91266.1 29010.8 | 11.5 | 7.2 11.8 | 48.8 11.9 | 5.0 3.1 | 46.7 | 471474.3 196.3 42163.0 | 0.0 0.0 | | 201 | 343302.8 90265.7 28450.7 | 10.8 | 7.2 10.0 | 30.2 20.0 | 6.3 2.1 | 30.7 | 468287.2 203.8 42157.7 | 0.0 0.0 | | 202 | 347252.0 90232.5 28338.9 | 10.1 | 7.5 10.0 | 40.3 19.8 | 5.2 0.7 | 58.7 | 472091.4 200.5 42157.0 | 0.0 0.0 | | 203 | 342763.8 90681.4 28693.0 | 10.5 | 7.1 10.0 | 42.3 13.9 | 5.7 1.9 | 36.8 | 468406.2 197.1 42158.7 | 0.0 0.0 | | 204 | 348227.2 90050.1 28818.0 | 10.6 | 7.6 11.8 | 36.3 20.3 | 9.0 3.1 | 63.0 | 473363.3 203.9 42161.3 | 0.0 0.0 | | 205 | 348326.9 93495.4 28974.1 | 11.2 | 8.0 10.7 | 41.6 20.8 | 7.9 0.2 | 58.4 | 477064.5 198.9 42154.5 | 0.0 0.0 | | 206 | 342000.0 91182.2 28027.8 | 10.0 | 7.1 10.2 | 33.7 18.9 | 5.1 0.0 | 30.6 | 467478.0 200.2 42163.2 | 0.0 0.0 | | 207 | 348307.8 90687.3 30741.6 | 10.3 | 7.8 10.3 | 39.0 27.3 | 9.7 0.9 | 42.7 | 476004.8 196.4 42159.8 | 0.0 0.0 | | 208 | 343268.6 91840.7 29740.8 | 12.7 | 7.3 11.9 | 35.2 18.0 | 9.3 3.4 | 31.5 | 471118.1 196.4 42155.0 | 0.0 0.0 | | 209 | 343998.0 93992.6 28295.0 | 10.8 | 7.4 11.2 | 39.7 20.9 | 5.8 4.0 | 67.5 | 472553.6 193.4 42154.7 | 2.6 0.0 | | 210 | 347468.6 93836.2 30606.9 | 12.6 | 7.6 11.8 | 42.7 29.2 | 9.1 2.2 | 62.0 | 478179.6 196.2 42158.0 | 0.0 0.0 | | 211 | 343667.0 91086.8 30190.8 | 11.7 | 7.1 11.0 | 39.0 27.4 | 5.9 1.2 | 42.6 | 471212.7 200.3 42168.3 | 0.0 0.0 | | 212 | 346480.5 89889.7 29733.2 | 10.6 | 7.5 10.5 | 40.0 18.4 | 7.0 1.2 | 30.6 | 472371.4 199.0 42174.1 | 0.0 0.0 | | 213 | 348098.9 90280.6 29796.3 | 13.7 | 7.5 10.7 | 42.9 17.2 | 6.1 3.4 | 48.4 | 474443.9 202.8 42159.2 | 0.0 0.0 | | 214 | 343120.8 89948.6 29123.3 | 10.5 | 7.1 11.3 | 37.2 19.3 | 5.3 1.4 | 30.7 | 468460.7 197.0 42158.8 | 0.0 0.0 | | 215 | 343653.3 91859.0 29531.8 | 10.2 | 7.3 10.9 | 43.9 16.4 | 8.1 0.5 | 33.2 | 471312.1 198.1 42156.5 | 0.0 0.0 | | 216 | 345959.6 89721.5 29378.0 | 12.1 | 7.4 10.0 | 40.9 15.0 | 5.0 2.4 | 30.1 | 471327.2 198.8 42164.8 | 0.0 0.0 | | 217 | 347822.0 91251.7 29887.2 | 12.1 | 7.3 10.3 | 46.8 26.1 | 5.1 4.0 | 70.1 | 475228.9 200.8 42160.6 | 0.0 0.0 | | 218 | 346083.5 89336.4 28319.4 | 11.4 | 7.1 10.6 | 31.9 28.7 | 5.1 1.1 | 77.4 | 470007.4 204.5 42174.0 | 0.5 0.0 | | 219 | 342796.1 89791.5 28832.9 | 10.8 | 7.1 10.3 | 35.7 19.4 | 6.1 0.6 | 36.5 | 467688.5 198.9 42156.9 | 0.0 0.0 | | 220 | 346452.2 89047.3 28243.1 | 10.3 | 7.5 10.6 | 30.1 20.3 | 5.9 1.2 | 38.2 | 470010.6 196.7 42165.3 | 0.0 0.0 | | 221 | 344212.7 90518.1 28227.4 | 11.1 | 7.1 10.4 | 37.7 19.7 | 6.4 1.0 | 63.7 | 469226.2 196.3 42171.9 | 0.0 0.0 | | 222 | 345483.0 89029.4 29149.1 | 12.3 | 7.2 10.1 | 40.2 10.6 | 8.6 3.9 | 38.1 | 469929.5 195.9 42155.8 | 0.1 0.0 | | 223 | 347680.5 89524.0 28712.6 | 12.0 | 7.6 10.3 | 34.6 17.1 | 5.0 3.5 | 36.9 | 472185.1 201.4 42169.1 | 0.0 0.0 | | 224 | 347904.1 89984.2 28000.0 | 12.1 | 7.5 10.5 | 36.3 18.8 | 5.2 3.2 | 70.2 | 472156.3 196.7 42161.4 | 0.0 0.0 | | 225 | 347765.8 89446.6 28472.0 | 14.0 | 7.1 10.8 | 32.7 27.2 | 10.0 3.6 | 111.4 | 471952.4 206.4 42176.3 | 2.4 1.8 | | 226 | 343409.7 89483.7 28591.8 | 10.0 | 7.0 11.7 | 33.7 23.0 | 9.2 1.2 | 63.4 | 467753.2 202.5 42156.2 | 0.0 0.0 | | 227 | 343654.0 90258.3 29685.0 | 10.4 | 7.2 10.0 | 42.9 17.1 | 7.0 1.9 | 34.5 | 469865.3 197.7 42165.6 | 0.0 0.0 | | 228 | 345815.3 89950.2 28071.1 | 11.5 | 7.2 10.7 | 35.1 23.1 | 5.3 1.7 | 75.5 | 470104.6 196.5 42154.6 | 0.0 0.0 | | 229 | 345014.9 90441.6 28590.3 | 10.7 | 7.1 11.3 | 40.4 23.9 | 6.1 1.2 | 72.3 | 470314.8 210.2 42164.7 | 6.2 0.0 | | 230 | 344590.8 89547.2 29968.0 | 10.9 | 7.2 10.7 | 33.9 26.6 | 8.1 0.8 | 44.3 | 470373.9 193.5 42152.7 | 2.5 1.8 | | 231 | 343882.4 91891.2 28638.0 | 10.2 | 7.5 10.4 | 35.5 17.5 | 8.0 1.8 | 35.0 | 470679.6 198.0 42170.6 | 0.0 0.0 | | 232 | 343430.0 91940.8 28226.3 | 11.4 | 7.2 11.2 | 36.3 19.8 | 6.1 2.0 | 57.5 | 469865.1 196.7 42154.8 | 0.0 0.0 | | 233 | 343987.1 92919.9 28236.0 | 10.4 | 7.4 10.6 | 34.3 24.5 | 6.8 1.3 | 63.1 | 471411.0 198.0 42171.7 | 0.0 0.0 |

195

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 234 | 344749.2 91991.2 28143.8 | 12.7 | 7.4 10.3 | 31.9 19.4 | 7.2 2.5 | 52.0 | 471152.3 203.8 42164.3 | 0.0 0.0 | | 235 | 344155.9 93692.2 30562.6 | 13.9 | 7.3 11.1 | 46.5 19.2 | 8.4 4.4 | 47.7 | 474678.7 200.0 42154.8 | 0.0 0.0 | | 236 | 344160.5 92865.6 28579.4 | 11.9 | 7.4 10.8 | 38.5 14.9 | 8.3 4.2 | 49.5 | 471873.5 197.1 42158.1 | 0.0 0.0 | | 237 | 345283.2 89425.9 28121.6 | 10.6 | 7.2 10.1 | 37.4 18.8 | 5.0 2.1 | 57.4 | 469098.8 202.9 42164.4 | 0.0 0.0 | | 238 | 347421.5 91371.4 28915.5 | 12.6 | 7.5 11.1 | 36.4 24.7 | 6.8 1.4 | 71.3 | 473976.4 198.0 42156.4 | 0.0 0.0 | | 239 | 346090.1 91429.0 30131.6 | 10.2 | 7.2 11.1 | 49.5 23.6 | 7.3 3.7 | 57.9 | 473918.6 197.6 42157.0 | 0.0 0.0 | | 240 | 345750.0 90649.9 28766.0 | 14.0 | 7.2 11.6 | 36.4 17.6 | 8.6 4.9 | 66.4 | 471434.0 209.3 42161.3 | 5.3 0.0 | | 241 | 344995.4 92584.1 28969.7 | 11.2 | 7.5 11.0 | 33.6 25.2 | 9.7 0.0 | 59.2 | 472817.2 202.3 42164.5 | 0.0 0.0 | | 242 | 344864.7 90012.7 29076.3 | 11.2 | 7.0 10.7 | 45.1 19.6 | 7.0 1.3 | 67.9 | 470221.7 200.8 42156.6 | 0.0 0.0 | | 243 | 345865.3 93392.6 28021.6 | 14.0 | 7.3 10.9 | 43.5 15.7 | 7.8 0.6 | 81.2 | 473547.5 197.8 42157.2 | 0.0 0.0 | | 244 | 344256.8 90249.9 28611.3 | 11.7 | 7.2 10.6 | 33.1 21.1 | 7.5 1.1 | 56.5 | 469386.1 187.2 42145.7 | 8.8 8.8 | | 245 | 346755.4 93814.6 29192.0 | 11.9 | 7.7 10.5 | 35.6 28.6 | 5.6 2.9 | 62.1 | 476030.0 202.1 42159.6 | 0.0 0.0 | | 246 | 347220.9 93182.1 28336.8 | 10.5 | 7.7 10.8 | 45.4 11.7 | 8.5 3.5 | 56.4 | 475007.9 203.3 42156.3 | 0.0 0.0 | | 247 | 343281.5 92844.2 29676.1 | 10.0 | 7.3 11.4 | 47.5 17.2 | 7.1 0.5 | 32.4 | 472069.8 200.0 42159.0 | 0.0 0.0 | | 248 | 342952.4 90070.3 28733.2 | 10.0 | 7.2 11.7 | 31.9 20.6 | 5.5 2.8 | 30.0 | 468023.9 196.2 42154.6 | 0.0 0.0 | | 249 | 347258.4 91800.2 29558.7 | 10.3 | 7.6 10.7 | 47.7 16.5 | 8.8 3.3 | 54.3 | 474885.3 204.0 42171.1 | 0.0 0.0 | | 250 | 343557.7 89720.0 28274.1 | 10.2 | 7.2 10.3 | 31.4 17.3 | 8.8 0.1 | 33.7 | 467819.8 198.8 42167.4 | 0.0 0.0 | | 251 | 344170.8 92974.6 28041.3 | 12.5 | 7.0 11.0 | 38.5 27.5 | 5.8 0.2 | 94.0 | 471454.7 200.0 42159.3 | 0.0 0.0 | | 252 | 344848.2 89913.9 28090.6 | 10.8 | 7.3 11.9 | 37.2 12.4 | 5.5 1.8 | 30.0 | 469120.7 196.0 42168.6 | 0.0 0.0 | | 253 | 346195.7 90805.5 28221.7 | 11.7 | 7.4 10.5 | 41.2 10.5 | 7.9 3.8 | 50.8 | 471490.9 201.0 42154.7 | 0.0 0.0 | | 254 | 343407.5 91304.7 29221.2 | 10.7 | 7.2 11.6 | 41.8 18.4 | 8.1 0.2 | 44.8 | 470201.3 200.0 42157.4 | 0.0 0.0 | | 255 | 347060.4 93117.4 28000.0 | 11.6 | 7.4 10.5 | 47.7 19.0 | 6.5 0.7 | 91.7 | 474445.8 198.7 42155.6 | 0.0 0.0 | | 256 | 345196.2 91472.4 29088.7 | 10.1 | 7.2 11.1 | 37.6 29.7 | 7.0 3.1 | 75.1 | 472025.3 203.8 42155.0 | 0.0 0.0 | | 257 | 343727.1 90548.4 29968.3 | 11.5 | 7.2 10.4 | 43.2 18.2 | 6.4 1.2 | 31.7 | 470511.8 198.8 42155.8 | 0.0 0.0 | | 258 | 347105.3 90075.0 28000.0 | 13.7 | 7.3 12.0 | 30.5 23.6 | 7.2 0.2 | 77.4 | 471448.3 199.2 42157.6 | 0.0 0.0 | | 259 | 345254.2 90430.9 29214.6 | 10.9 | 7.2 10.9 | 40.9 21.3 | 5.4 3.3 | 54.5 | 471167.7 202.8 42157.5 | 0.0 0.0 | | 260 | 347298.2 89549.6 28991.4 | 10.8 | 7.2 10.6 | 42.4 25.2 | 6.4 1.0 | 79.4 | 472107.3 198.1 42157.1 | 0.0 0.0 | | 261 | 345841.4 91218.4 28123.1 | 11.6 | 7.6 10.5 | 32.7 13.4 | 8.0 3.2 | 31.0 | 471450.9 196.9 42157.7 | 0.0 0.0 | | 262 | 343579.3 91077.9 28669.4 | 14.0 | 7.1 10.8 | 32.1 21.6 | 6.4 0.2 | 47.8 | 469594.6 199.6 42165.1 | 0.0 0.0 | | 263 | 346533.7 93129.9 29541.8 | 11.9 | 7.8 11.6 | 41.6 18.6 | 7.8 2.5 | 45.5 | 475473.3 197.2 42161.7 | 0.0 0.0 | | 264 | 347721.7 93909.7 30355.1 | 13.2 | 7.5 10.7 | 43.8 29.5 | 6.7 4.7 | 69.2 | 478254.5 195.9 42154.8 | 0.1 0.0 | | 265 | 347419.3 89825.1 28622.5 | 13.7 | 7.0 11.9 | 43.2 22.6 | 9.9 1.0 | 112.7 | 472134.9 197.7 42157.6 | 0.0 0.0 | | 266 | 342856.7 90078.9 28600.2 | 10.7 | 7.1 11.2 | 32.2 17.5 | 9.5 3.3 | 42.7 | 467803.8 198.5 42154.8 | 0.0 0.0 | | 267 | 344143.6 91416.0 28986.5 | 10.4 | 7.3 10.2 | 38.2 23.0 | 6.7 0.8 | 53.3 | 470814.1 196.0 42154.5 | 0.0 0.0 |

196

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 268 | 347243.5 92080.8 28296.2 | 13.4 | 7.4 11.3 | 48.0 10.1 | 5.2 1.5 | 57.8 | 473888.5 196.1 42155.1 | 0.0 0.0 | | 269 | 347448.2 90949.5 28795.0 | 10.7 | 7.4 10.7 | 43.7 21.8 | 5.3 0.8 | 67.7 | 473460.8 195.6 42160.0 | 0.4 0.0 | | 270 | 345932.9 93447.8 29572.7 | 13.7 | 7.7 11.9 | 43.6 14.5 | 7.5 0.4 | 30.8 | 475221.4 202.7 42154.6 | 0.0 0.0 | | 271 | 347733.7 93629.4 30893.3 | 12.6 | 7.5 11.9 | 49.5 26.1 | 9.9 1.8 | 56.9 | 478524.4 205.4 42175.9 | 1.4 1.4 | | 272 | 343317.1 90857.0 28193.5 | 10.8 | 7.2 10.7 | 39.1 14.3 | 5.9 1.2 | 38.7 | 468635.7 200.3 42155.9 | 0.0 0.0 | | 273 | 343672.9 90137.2 28078.2 | 10.8 | 7.1 10.0 | 30.8 23.4 | 6.4 0.9 | 58.5 | 468156.3 197.5 42154.9 | 0.0 0.0 | | 274 | 342654.5 91320.3 28305.7 | 10.5 | 7.1 10.6 | 40.5 17.3 | 5.0 2.7 | 51.5 | 468548.5 199.3 42173.5 | 0.0 0.0 | | 275 | 344868.7 90823.6 29877.8 | 10.2 | 7.2 11.2 | 40.4 26.1 | 9.7 1.7 | 63.4 | 471838.0 196.2 42166.1 | 0.0 0.0 | | 276 | 343019.3 91732.2 29089.1 | 10.6 | 7.2 10.9 | 30.0 28.8 | 5.7 2.7 | 46.3 | 470108.7 203.1 42174.4 | 0.0 0.0 | | 277 | 346575.9 92651.2 28197.6 | 11.3 | 7.3 11.4 | 45.4 21.1 | 7.2 3.8 | 99.0 | 473692.7 198.8 42155.6 | 0.0 0.0 | | 278 | 343768.6 90146.0 28384.1 | 12.9 | 7.2 11.0 | 30.0 18.8 | 5.0 1.2 | 30.0 | 468566.6 196.1 42166.1 | 0.0 0.0 | | 279 | 346098.8 92948.1 28198.0 | 11.9 | 7.1 11.9 | 43.6 25.5 | 6.6 4.4 | 110.1 | 473512.9 197.2 42172.1 | 0.0 0.0 | | 280 | 348921.8 90017.1 29200.5 | 12.0 | 7.8 11.5 | 35.3 20.2 | 5.0 0.1 | 30.5 | 474407.4 197.1 42157.0 | 0.0 0.0 | | 281 | 346786.0 92596.5 28263.9 | 13.0 | 7.8 10.8 | 37.3 11.0 | 7.6 2.0 | 30.7 | 473914.4 197.3 42155.8 | 0.0 0.0 | | 282 | 343886.4 90274.5 28178.6 | 10.3 | 7.2 10.4 | 37.9 14.1 | 7.7 4.0 | 48.8 | 468607.5 202.5 42154.8 | 0.0 0.0 | | 283 | 345841.1 91596.5 29861.3 | 12.5 | 7.6 11.3 | 30.8 23.4 | 8.6 4.6 | 38.3 | 473566.8 197.1 42172.2 | 0.0 0.0 | | 284 | 348913.7 93460.3 28769.0 | 11.8 | 7.4 11.7 | 45.4 28.9 | 9.9 3.5 | 116.2 | 477410.9 196.7 42156.4 | 0.0 0.0 | | 285 | 347085.9 93079.3 28157.0 | 10.4 | 7.5 10.0 | 36.8 29.3 | 8.5 0.5 | 96.5 | 474590.2 198.4 42160.4 | 0.0 0.0 | | 286 | 346703.4 91723.5 29645.9 | 10.8 | 7.6 10.9 | 34.8 28.8 | 7.6 0.0 | 51.8 | 474340.9 196.1 42156.1 | 0.0 0.0 | | 287 | 343904.4 90538.5 29352.1 | 13.2 | 7.1 10.7 | 35.1 21.1 | 6.5 1.4 | 42.3 | 470063.0 198.5 42157.1 | 0.0 0.0 | | 288 | 347281.0 90995.1 29962.4 | 11.7 | 7.6 10.0 | 41.1 20.0 | 8.8 3.7 | 52.7 | 474506.6 196.5 42159.7 | 0.0 0.0 | | 289 | 347533.2 89336.9 28304.7 | 10.1 | 7.4 10.6 | 31.5 26.0 | 5.0 2.4 | 65.4 | 471442.8 197.3 42156.3 | 0.0 0.0 | | 290 | 348642.9 89492.7 30634.4 | 11.4 | 7.5 10.8 | 40.7 26.7 | 5.5 3.2 | 42.8 | 475037.9 200.6 42172.7 | 0.0 0.0 | | 291 | 347543.0 91464.0 30440.4 | 13.5 | 7.6 11.2 | 33.0 29.2 | 6.4 2.6 | 48.9 | 475715.3 177.0 42131.9 | 19.0 22.6 | | 292 | 343418.6 91664.2 29885.8 | 10.9 | 7.1 10.5 | 45.3 21.8 | 9.0 2.0 | 59.0 | 471236.6 198.5 42154.6 | 0.0 0.0 | | 293 | 345659.0 92088.4 28773.1 | 13.7 | 7.5 10.9 | 34.0 20.5 | 5.8 2.0 | 46.1 | 472788.5 200.3 42162.3 | 0.0 0.0 | | 294 | 347924.9 90162.6 28370.9 | 10.4 | 7.6 11.1 | 33.2 23.3 | 5.2 4.6 | 64.0 | 472726.3 197.8 42171.0 | 0.0 0.0 | | 295 | 347321.9 89000.0 28542.0 | 12.8 | 7.4 10.0 | 39.3 11.0 | 7.2 0.4 | 37.0 | 471132.0 198.7 42167.6 | 0.0 0.0 | | 296 | 344605.3 90145.8 28012.1 | 11.2 | 7.2 10.8 | 30.2 23.2 | 5.4 1.7 | 56.4 | 469031.2 199.5 42155.5 | 0.0 0.0 | | 297 | 346074.7 93291.2 30451.7 | 13.0 | 7.4 11.6 | 45.5 24.6 | 9.9 0.7 | 59.2 | 476085.6 202.8 42160.7 | 0.0 0.0 | | 298 | 348086.8 91647.1 28669.0 | 11.4 | 7.5 10.3 | 45.3 19.1 | 6.8 0.4 | 74.2 | 474670.8 193.5 42152.1 | 2.5 2.4 | | 299 | 345244.6 90130.6 28726.6 | 11.5 | 7.3 11.0 | 35.7 17.4 | 7.5 2.8 | 47.8 | 470369.9 196.6 42164.2 | 0.0 0.0 | | 300 | 347551.6 91978.9 29523.2 | 10.3 | 7.8 10.7 | 42.1 20.1 | 6.6 3.5 | 46.9 | 475321.7 196.5 42157.1 | 0.0 0.0 | | 301 | 346768.4 89210.9 29210.0 | 10.4 | 7.5 10.9 | 42.2 12.0 | 7.9 1.3 | 30.0 | 471457.2 196.2 42156.8 | 0.0 0.0 |

197

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 302 | 346645.8 90965.4 28714.6 | 11.6 | 7.6 10.8 | 40.3 11.8 | 5.9 4.7 | 34.7 | 472593.9 196.2 42154.8 | 0.0 0.0 | | 303 | 345345.5 90895.9 29221.8 | 11.7 | 7.1 10.6 | 44.2 22.1 | 5.8 2.5 | 69.9 | 471731.2 196.3 42168.8 | 0.0 0.0 | | 304 | 348154.3 90269.6 29948.3 | 10.1 | 7.2 11.2 | 47.4 29.8 | 9.8 0.0 | 76.1 | 474640.1 199.0 42163.4 | 0.0 0.0 | | 305 | 343763.8 89710.4 30178.0 | 10.5 | 7.1 11.0 | 46.4 16.5 | 7.1 1.4 | 30.9 | 469920.3 196.2 42155.7 | 0.0 0.0 | | 306 | 345141.2 89303.6 28520.3 | 10.9 | 7.0 10.4 | 36.2 24.4 | 5.9 2.8 | 75.5 | 469233.0 202.2 42158.7 | 0.0 0.0 | | 307 | 348360.6 89001.2 28308.8 | 12.2 | 7.3 10.6 | 42.9 11.9 | 5.5 4.5 | 57.3 | 471938.6 217.2 42183.5 | 13.2 9.0 | | 308 | 342401.9 92136.8 28313.9 | 11.4 | 7.0 10.6 | 33.4 27.1 | 8.1 0.1 | 83.4 | 469120.5 180.8 42165.3 | 15.2 0.0 | | 309 | 345666.0 89239.6 30502.1 | 12.0 | 7.2 11.1 | 38.6 25.2 | 6.8 4.4 | 59.5 | 471675.8 162.1 42120.6 | 33.9 33.9 | | 310 | 347751.7 91562.2 30119.3 | 12.9 | 7.6 11.8 | 37.8 26.5 | 8.1 2.8 | 60.9 | 475701.1 198.0 42160.5 | 0.0 0.0 | | 311 | 343547.3 91537.2 28621.4 | 10.2 | 7.0 11.0 | 47.2 15.2 | 7.1 1.5 | 59.5 | 469973.9 201.5 42155.8 | 0.0 0.0 | | 312 | 347643.5 93065.9 29529.4 | 11.7 | 7.7 10.6 | 37.4 29.0 | 5.9 0.1 | 53.2 | 476506.8 203.5 42157.4 | 0.0 0.0 | | 313 | 342991.7 91323.4 28785.5 | 10.1 | 7.2 10.2 | 30.9 25.5 | 6.3 0.8 | 40.2 | 469368.6 199.5 42169.7 | 0.0 0.0 | | 314 | 343691.0 91251.9 28340.0 | 11.9 | 7.0 10.1 | 31.4 27.5 | 5.3 1.8 | 69.6 | 469550.9 202.6 42159.7 | 0.0 0.0 | | 315 | 345917.3 91057.5 28876.5 | 11.8 | 7.3 11.4 | 41.2 18.4 | 9.1 4.6 | 76.3 | 472119.2 202.6 42159.8 | 0.0 0.0 | | 316 | 344876.7 91990.9 30534.2 | 12.7 | 7.2 10.1 | 44.7 23.5 | 7.5 2.2 | 49.6 | 473669.7 196.1 42154.5 | 0.0 0.0 | | 317 | 343700.9 92715.1 29154.3 | 11.4 | 7.4 11.8 | 37.1 20.4 | 9.9 3.3 | 56.3 | 471838.4 199.0 42163.2 | 0.0 0.0 | | 318 | 343480.1 89319.3 28069.3 | 12.0 | 7.1 10.2 | 32.3 15.3 | 6.2 0.2 | 30.0 | 467136.6 198.4 42159.5 | 0.0 0.0 | | 319 | 345210.7 92442.7 28592.7 | 12.2 | 7.4 10.8 | 36.1 23.2 | 6.5 0.7 | 60.9 | 472514.1 202.4 42162.0 | 0.0 0.0 | | 320 | 345006.9 92295.7 29816.2 | 13.0 | 7.5 12.0 | 43.2 13.9 | 9.3 0.4 | 31.6 | 473386.8 196.3 42158.7 | 0.0 0.0 | | 321 | 345756.2 91933.5 28942.6 | 10.3 | 7.7 10.0 | 38.5 14.0 | 9.3 4.1 | 35.4 | 472900.3 201.0 42157.6 | 0.0 0.0 | | 322 | 347060.5 92185.8 29649.9 | 13.9 | 7.4 11.9 | 37.7 25.8 | 8.4 4.7 | 80.1 | 475164.2 194.3 42153.0 | 1.7 1.5 | | 323 | 347200.7 90309.5 28102.7 | 10.4 | 7.4 10.2 | 44.2 14.3 | 5.2 0.9 | 56.3 | 471880.9 202.1 42155.9 | 0.0 0.0 | | 324 | 345345.5 91680.2 28273.1 | 13.3 | 7.1 11.4 | 45.7 16.2 | 5.2 0.3 | 73.2 | 471566.8 196.2 42160.8 | 0.0 0.0 | | 325 | 347415.0 93198.9 29928.5 | 12.5 | 7.6 10.5 | 41.5 28.2 | 9.1 1.5 | 73.9 | 476810.5 196.3 42154.6 | 0.0 0.0 | | 326 | 343132.8 92476.1 28557.9 | 12.2 | 7.3 10.5 | 34.0 18.0 | 5.0 4.8 | 36.6 | 470434.8 196.3 42161.7 | 0.0 0.0 | | 327 | 342501.1 90325.4 28507.4 | 10.5 | 7.0 11.8 | 30.9 23.2 | 5.5 3.4 | 45.2 | 467601.9 203.6 42165.0 | 0.0 0.0 | | 328 | 344986.2 89118.0 28154.8 | 10.5 | 7.2 11.8 | 31.3 20.7 | 5.8 0.9 | 45.2 | 468527.0 196.9 42156.9 | 0.0 0.0 | | 329 | 345496.0 89312.4 28754.6 | 12.5 | 7.2 11.7 | 31.0 20.9 | 6.2 3.5 | 48.1 | 469831.1 200.8 42156.7 | 0.0 0.0 | | 330 | 345608.0 90245.0 29046.5 | 10.9 | 7.3 11.1 | 31.7 26.8 | 10.0 0.3 | 68.2 | 471167.5 194.2 42155.0 | 1.8 0.0 | | 331 | 348192.8 93393.2 29452.9 | 13.9 | 7.6 10.5 | 44.1 23.9 | 7.6 0.1 | 73.1 | 477306.8 201.9 42155.8 | 0.0 0.0 | | 332 | 347039.4 91114.8 30405.6 | 11.6 | 7.4 11.4 | 44.7 24.6 | 8.3 3.7 | 59.7 | 474827.7 203.4 42168.2 | 0.0 0.0 | | 333 | 348240.4 89893.0 29000.4 | 12.6 | 7.4 10.3 | 40.6 17.1 | 10.0 4.1 | 76.4 | 473401.8 203.2 42157.2 | 0.0 0.0 | | 334 | 344576.6 89288.4 28505.3 | 10.8 | 7.1 10.2 | 37.2 16.6 | 8.7 1.9 | 56.4 | 468638.3 198.1 42166.5 | 0.0 0.0 | | 335 | 346637.5 92765.5 29297.5 | 12.2 | 7.3 11.0 | 45.5 22.4 | 10.0 4.8 | 93.5 | 474968.6 195.5 42154.0 | 0.5 0.5 |

198

Appendix B3 Sample TRAJ2DF Data Files

Traj2dF - Evolution Run Results

Case Study: Hypothetical Launch Vehicle

| Run | S1Mass S2Mass S2Mass | S1Eps|S1VRT S1POA|S2IPA S2FPA|S3IPA S3FPA| S23ICT| GLOW Hp Ra | Con1 Con2 | | | kg kg kg | | sec deg | deg deg | deg deg | sec | kg km km | | | 336 | 343861.2 90191.5 29025.6 | 10.4 | 7.2 10.0 | 40.9 15.7 | 7.6 1.8 | 42.2 | 469346.3 196.0 42154.5 | 0.0 0.0 | | 337 | 347641.2 90645.5 28867.7 | 12.2 | 7.3 10.3 | 44.1 21.4 | 6.9 0.2 | 78.7 | 473422.4 200.9 42168.6 | 0.0 0.0 | | 338 | 348878.8 91297.8 29404.0 | 12.0 | 7.7 11.8 | 48.0 13.9 | 8.5 1.3 | 55.3 | 475848.5 202.5 42156.3 | 0.0 0.0 | | 339 | 347485.4 93726.5 30691.5 | 13.3 | 7.7 11.5 | 40.7 29.2 | 8.3 2.9 | 55.4 | 478171.4 203.2 42167.4 | 0.0 0.0 | | 340 | 348750.1 90706.9 30154.6 | 12.2 | 7.6 11.8 | 44.0 22.7 | 5.3 1.6 | 46.7 | 475879.6 197.2 42156.5 | 0.0 0.0 | | 341 | 345753.6 92869.9 28208.2 | 13.2 | 7.6 10.9 | 35.2 17.6 | 6.5 1.6 | 50.7 | 473099.7 197.2 42159.5 | 0.0 0.0 | | 342 | 342128.5 91749.7 28188.4 | 11.7 | 7.0 11.5 | 41.2 14.3 | 5.2 2.7 | 48.6 | 468334.7 201.6 42171.8 | 0.0 0.0 | | 343 | 348783.2 89468.4 28883.9 | 10.2 | 7.6 10.4 | 33.0 24.6 | 7.8 2.9 | 64.4 | 473403.5 197.3 42163.1 | 0.0 0.0 | | 344 | 347728.1 89549.0 30676.8 | 13.3 | 7.3 11.0 | 47.4 17.4 | 9.4 4.4 | 58.4 | 474222.0 196.4 42154.6 | 0.0 0.0 | | 345 | 345884.5 92966.7 29459.5 | 10.7 | 7.6 10.9 | 41.3 24.9 | 7.7 1.5 | 62.3 | 474578.6 188.7 42147.2 | 7.3 7.3 | | 346 | 347314.3 91318.3 30483.5 | 13.1 | 7.3 11.2 | 45.4 25.0 | 7.8 1.2 | 58.8 | 475384.1 202.1 42156.0 | 0.0 0.0 | | 347 | 344692.2 91836.6 28180.7 | 10.0 | 7.4 10.0 | 41.0 11.0 | 9.5 3.3 | 45.4 | 470977.6 203.8 42164.1 | 0.0 0.0 | | 348 | 346479.9 90953.4 29955.3 | 10.2 | 7.6 10.5 | 41.5 19.9 | 7.4 1.9 | 34.0 | 473656.6 196.1 42155.3 | 0.0 0.0 | | 349 | 345111.6 90371.6 28433.4 | 12.6 | 7.0 10.6 | 35.9 22.9 | 8.6 4.9 | 95.0 | 470184.6 197.1 42165.5 | 0.0 0.0 | | 350 | 344103.0 93776.6 29852.7 | 12.1 | 7.5 10.0 | 40.2 22.6 | 8.9 0.5 | 45.6 | 474000.3 197.1 42169.9 | 0.0 0.0 |

Ú And so on

199

Appendix C Sample STAGEX Spreadsheet

Appendix C Sample STAGEX Spreadsheet

200

Appendix C Sample STAGEX Spreadsheet

The STAGEX Spreadsheet This is the final version of STAGEX. A Solver routine was added to do the GLOW minimisation automatically. Delta-V reserves are included in all stages. The ability to specify the number of engines in the first and second stages was added along with constraints for the limits of the lift-off thrust to weight, the S1 burnout axial acceleration and the S2 burnout axial acceleration. Provision for the growth of S1 thrust from sea level to vacuum is made as well as the assumption of engine commonality between S1 and S2 being included implicitly with an allowance for Ivac increase between S1 and S2 due to a larger nozzle expansion ratio skirt being utilised on the S2 engine. The example spreadsheet on the following page shows design calculations for an all LOX/LH2 three stage vehicle with the same payload capability and characteristic velocity as an AR44L vehicle.

201

Appendix C Sample STAGEX Spreadsheet

202